(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Transactions of the Royal Society of Edinburgh"

TRANSACTIONS 



OF THE 



KOYAL SOCIETY OF EDINBURGH. 



S.ii-.C.'so 



TRANSACTIONS 



OP THE 



ROYAL SOCIETY 



OF 



EDINBURGH. 



VOL. XLIX. 




EDINBURGH : 

PUBLISHED BY ROBEET GEANT & SON, 107 PEINCES STEEET, 
AND WILLIAMS & NOEGATE, 14 HENEIETTA STEEET, COVENT GAEDEN, LONDON. 



MDCCCCXIV. 



^0. I. Published January 1913. 


No. XI. 


Published 


September 1, 1913. 


n. 


May 23, 1913. 


XII. 


n 


September 20, 1913 


III. 


June 9, 1913. 


XIII. 


ji 


October 17, 1913. 


TV. 


June 27, 1913. 


XIV. 


>) 


November 7, 1913. 


V. 


June 24, 1913. 


XV. 


>> 


December 2, 1913. 


VI. 


July 24, 1913. 


XVI. 


)j 


November 18, 1913 


VII. 


July 4, 1913. 


XVII. 


)i 


January 19, 1914. 


VIII. 


August 4, 1913. 


XVIII. 


)> 


February 24, 1914. 


IX. 


August 20, 1913. 


XIX. 


., 


March 30, 1914. 


X. 


August 29, 1913. 









CONTENTS. 



PART I. (1912-13.) 

MBER 

I. Experimental Researches on the Specific Gravity and the Displace- 
ment of some Saline Solutions. By J. Y. Buchanan, jVI.A., F.R.S., 
Commandeur de I'Ordre de St Charles de Monaco, Chemist and 
Physicist of the Challenger Expedition, Vice-President du Comit6 
de Perfectionnement de I'lnstitut Oceanographique (Fondation 
Albert l®^ Prince de Monaco), ..... 



PART 11. (1912-13.) 

II. The Antarctic Fishes of the Scottish National Antarctic Expedition. 
By C. Tate Regan, M.A., Assistant in the British Museum (Natural 
History). (With Eleven Plates and Six Text-Figs.), . . 229 

in. A Monograph on the general Morphology of the Myxinoid Fishes, 
based on a study of Myxine. Part V. : The Anatomy of the Gut 
and its Appendages. By F. J. Cole, D.Sc. Oxon., Professor of 
Zoology, University College, Reading. (With Four Plates), . 293 

IV. On the Skulls of Antarctic Seals : Scottish National Antarctic 
Expedition. By Willtam S. Bruce, LL.D., Director of the Scottish 
Oceanographical Laboratory. (With Five Plates), . . . 345 

V. PolychsBta of the Families Serpididae and Sabellidx, collected by the 
Scottish National Antarctic Expedition. By Helen L. M. Pixell, 
B.Sc, P.Z.8., Demonstrator of Zoology and Reid Fellow, Bedford 
College, University of London. (With One Plate), . . . 347 

VI. Caradocian Cystidea from Girvan. By F. A. Bather, M. A., D.Sc, 
F.R.S. (With Six Plates.) [Published by permission of the 
Trustees of the British Museum], . . . , .359 



vi CONTENTS. 

PART III. (1912-13.) 

NUMBKU PAGE 

VII. The Pterohranchia of the Scottish National Antarctic Expedition 
{1902 to 1904). By S. F. Harmer, Sc.D., F.R.S., Keeper of the 
Department of Zoology in the British Museum ; and W. G. Ride- 
wood, D.Sc, Lecturer on Biology in the Medical School of St Mary's 
Hospital, University of London. (With Five Text-Figures and Two 
Plates), . . . . . . . .531 

VIII. Measurements and Weights of Antarctic Seals taken hy the Scottish 
National Afitarctic Expeditio7i. By William S. Bruce, LL.D., 
F.R.S.E., Director of the Scottish Oceanographical Laboratory, 
Edinburgh. (With One Text-Diagram and Two Plates), . . 567 

IX. Spongiaires de r Expedition Antarctique Nationale Ecossaise. Par 
Emile Topsent, Professeur a I'Universite de Dijon. (Avec six 
planches), ........ 579 

X. Scottish National Antarctic Expedition, 1902-04 •' Deep-Sea Deposits. 
By J. H. Harvey Pirie, B.Sc, M.D., Geologist and Surgeon to the 
Expedition. (With a Map), . . . . . 645 

XL The Corals of the Scottish National Antarctic Expedition. By 
J. Stanley Gardiner, M.A., F.R.S., Professor of Zoology and Com- 
parative Anatomy in the University of Cambridge, . . .687 

XII. Studies in Floral Zygomorphy. I. The Initiation of Staminal 
Zygomorphy. By John McLean Thompson, M.A., B.Sc, Assistant 
in Botany, University of Glasgow. (With Two Plates), . .691 

XIII. Contributions to the Craniology of the People of the Empire of 

India. Part IV. : Bhils, Frontier Tribes of Burma, Pakdkku 
Tribes, South Shan Tribes, Tibetans. By Principal Sir William 
Turner, K.C.B., D.C.L., F.R.8., President of the Society, Knight of 
the Royal Prussian Order Pour le Merite. (With Three Plates and 
Figures in Text), . . . . . . T05 

XIV. On Intestinal Respiration in Annelids; with Considerations on the 

Origin and Evolution of the Vascular System in that Group. 
By J. Stephenson, M.B., D.Sc. (Lond.) ; Major, Indian Medical 
Service ; Professor of Biology, Government College, Lahore, . 735 



CONTENTS. vn 

PAET IV. (1913-14.) 

NUMBER PAGE 

XV. Glaciology of the South Orkneys : Scottish National Antarctic 
Expedition. By J. H. Harvey Pirie, B.Sc, M.D., Geologist and 
Surgeon. (With Fourteen Text-Figs, and Eleven Plates), . .831 

XVI, The Schizopoda, Stomatopoda, and non-Antarctic Isopoda of the 
Scottish National Antarctic Expedition. By AV alter M. Tattersall, 
D.Sc, Keeper of the Manchester Museum. (With One Plate), . 865 

XVII. An Analytical Theory of the Equilibrium of an Isotropic Elastic Rod 

of Circular Section. By John Dougall, M.A., D.Sc, . .895 

XVIII. The Chsetognatha of the Scottish National Antarctic Expedition of 

1902-1904-. By A. Pringlb Jameson, B.Sc. . . . 979 

XIX. Foraminifera, of the Scottish Natiojial Antao^ctic Expedition. By 
F. Gordon Pearcey, Bristol Museum ; late of the Challenger 
Expedition and Commission. (With Two Plates), . . .991 

Index, .......... 1045 



/ 









.WWWU.Iv/IL 



x^ 



TRANSACTIONS 



OF THE 



ROYAL SOCIETY OF EDINBURGH. 

VOLUME XLIX. PART L— SESSION 1912-13. 



CONTENTS. 

PAQB 

1. Experimental Researches on the Specific Gravity and the Displacement of 

some Saline Solutions. By J. Y. Buchanan, F.R.S.. . . . 1 




EDINBURGH: 

PUBLISHED BY ROBKET GRANT & SON, 107 PRINCES STREET, 

AND WILLIAMS & NORGATE, 14 HENRIETTA STRLET, COVENT GARDEN, LONDON. 



MDCCCCXII. 

Pi-ice Seven Shillings and Sixpence. 



TRANSACTIONS. 



I. — Experimental Researches on the Specific Gravity and the Displacement 
of some Saline Solutions. By J. Y. Buchanan, F.R.S. 

TABLE OF CONTENTS. 
SECTION I. 
Introduction. 

PAR. PAGE 

1. The Principles of Archimedes. Tliey embody the fundamental principles of the hydro- 17 

meter. Archimedes considered the immersion of a body in only one fluid, but the 
principles hold good when it is immersed in more than one fluid. 

2. Hydrometer suitable for Demonstrations on the Lecture Table. It was constructed originally 17 

in the year 1871 for use in tutorial classes in the University of Edinburgh, and especially 
to exhibit the determination of the specific gravity of solids lighter as well as heavier 
than water. A remarkable feature of the instrument is that no determination of 
weight is required either in its construction or its use. The only measurements made 
are those of length. 

3. Usefulness of the Hydrometer in the Study of Mineral Waters. It suggested itself while 19 

working as student and assistant of Fresenius (1863-1867), and later as Chemist and 
Physicist of the C/iallenger Expedition, during which it was used in investigating some 
mineral waters in the Philippine Islands. 

4. The Hydrometer in the " Challenger " Expedition. Early preparations for work in the 20 

expedition which lasted three and a half years. Indirect methods rejected. Adoption 
of the hydrometric method for determining the specific gravity of the water of 
the ocean. 

5. In designing the hydrometer, units in the fourth place of decimals were to be exact, and 20 

the exactness to be pushed as far as possible into the fifth place. Multiple sets of 
hydrometers rejected. One suitable hydrometer was designed, and its weight could 
be altered by the addition of accessory weights. 

6. The series of accessory weights prepared enabled the determination of the densities of all 21 

sea-waters, up to and including that of the Red Sea, to be made with the same glass 
hydrometer. But in the design of the series only single observations were contem- 
plated. Duplicate observations were occasionally obtained. The volume of the 
hydrometer was determined by floating it in distilled water at different temperatures. 
TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 1 



MR J. Y. BUCHANAN ON THE 

PAR. PAGE 

Full specification is given of the hydrometer and the accessory weights which were 
used in all the determinations made during the voyage of the Challenger. 

7. Importance of constancy of temperature in the practice of the hydrometric method. This 23 

was secured in the Challenger by her construction, and by the climate of the seas in which 
she cruised. The laboratory door was always locked while specific gravity observa- 
tions were being made. Table VIII. gives the instances of duplicate observations on 
identical samples of water with the same hydrometer differently weighted made in situ 
during the voyage. In voyages later than that of the Challenger hydrometric observa- 
tions were usually made in triplicate, and, later still, in series of nine. 

8. Quotation from the Report of the Sixth International Geographical Congress held in 25 

London in 1895, dealing with misunderstandings regarding the principle of the instru- 
ment and the qualifications required for the successful practice of the hydrometric 
method. 

SECTION II. 
The Principle and Construction of the Closed Hyrdrometer. 

9. Experiments assumed to be made in vacuo at the sea-level in lat. 45°. True weight of the 26 

hydrometer determined and its approximate volume determined by immersion in distilled 
water. Determination of weight of hydrometer in air and allowance for buoyancy 
of hydrometer when being weighed, and of exposed stem when floating in the 
liquid. 

10. Developments of the above for multiple observations. 27 

11. Preparation of Accessory Weights. These weights are made of wire, the heavier of brass 29 

and the lighter of aluminium. The pressure which they exert on the hydrometer, 
when in use, is due to their weight in air. 

12. Exposed Stem. The effect of buoyancy of the exposed stem is, in practice, almost 30 

inappreciable. 

13. Final Determination of the Weight of the Hydrometer. This is effected at different dates 31 

and under different meteorological conditions, from which the true weight in vacua 
is obtained. A table gives the exact weights in vacuo of hydrometers Nos. 17, 
21 and 3. 

14r. Experiments for the determination of the displacement of the hydrometer in distilled water, 33 

and description of Tables, A, in which these results are recorded. 

15. Correction for the Departure of the Mean Reading from 50 mm. 35 

16. Correction for Temperature. 35 

17. Tables A^ to Aj give details of the determination of the displacement in distilled water of 36 

hydrometer No. 17 at 15°, 19-5°, 23° and 26° C. ; and of hydrometer No. 21 at 19-5° C. 
Table B gives the observed weights of hydrometers Nos. 17, 21 and 3 when floating at 
the 50-mm. mark in distilled water at various temperatures. 



SECTION III. 

Determination of the Specific Gravity of a Saline Solution. 

18. Details of tliree series of observations made with hydrometer No. 17 in a solution of 1/8 CsCl 44 

in 1000 grams of water at 19'5° C. are given in Table C, which is arranged in the same 
manner as Tables A. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 3 

PAR. PAGE 

19. Influence of the Meniscus. The weight which causes the immersion of the hydrometer is 46 

its own weight plus that of the liquid meniscus which it carries on the stem above the 
line of flotation. This holds good when the hydrometer is floating in the solution and 
in distilled water alike. Each meniscus exerts the pressure of its own weight and con- 
tributes to the displacing weight of the instrument when it floats in the distilled water 
and the solution respectively. Data regarding the surface-tension of distilled water and 
of saline solutions have been taken from Tables 124 to 129 of the Smithsonian Collec- 
tion of Physical Tables, 5th edition, 1910. In a table, in this paragraph, the e0"ect of 
introducing the weight of the meniscus in the hydrometric determination of the specific 
gravity of some solutions has been calculated, with the result that, for a solution of NaCl 
having a specific gravity of 1 'OSS when the weights of both meniscuses are disregarded, 
the apparent specific gravity is increased by I '3 in the fifth place of decimals when 
efl'ect is given to the weight of both meniscuses. .As the density of average oceanic 
water is not greater than 1'027, the influence of the meniscus on its density would be 
represented, at the most, by 1 in the fifth place. Therefore my practice, adopted at 
the beginning, of disregarding the influence of meniscus on the density of sea-water, as 
determined by my hydrometric method, is justified, 

20. Serial Determination of the Specific Gravity of a Saline Solution. The table gives the 48 

specific gravity of 1/32 RbCl in 1000 grams of water at 19 '5° C. In it the specific gravity 
is deduced from each individual observation of four series of nine observations each, 
making in all thirty-six independent determinations, and they agree well with each other. 
The solutions of twenty-seven salts form the subject of the tables in Section V. They 
are KCl, RbCl, CsCl, KBr, RbBr, CsBr, KI, Rbl, Csl, which form the ennead* having 
the general formula MR; KCIO3, RbClOg, CsClOg, KBrOj, RbBrOg, CsBrOg, KIO3, 
RblOg, CsIOg, which form the ennead having the general formula MRO3 ; and NaCl, 
KNO3, RbNOg, CsNOg, LiN03, NaNOj, Sr(N03)2, Ba(N03)2 and Pb(N03)2. The 
concentrations of these solutions vary from 1/2 to 1/1024 gram-molecule of salt per 
1000 grams of water. In the cases of strong solutions the concentrations were from 
1 gram-molecule per 1000 grams of water upwards. 

21. Statistics relating to the Range of Variation of Temperature during a Series of Observations. 51 

The table gives statistics of the variations of the temperature of the liquid while a total 
of 1316 series, of nine observations each, was made with hydrometers Nos. 17 and 21, 
namely, 837 with No. 17 and 479 with No. 21. There was no sensible variation of 
temperature in 68 per cent, of those made with No. 17, and in 55 '2 per cent, of those 
made with No. 21. Considering the series made with both hydrometers for which the 
variation of temperature was not greater than 0"05° C, the percentages are almost 
identical, namely, 89-5 for No. 17 and 89"2 for No. 21. The maximum departure of 
the mean temperature from the standard (T), during any single series of observations, 
was 0'12° C. ; the mean departure was0'0075° C. The maximum range of temperature 
while a series of nine observations was being made was 0"30° C. ; the mean range cf 
temperature for the 1316 series, of nine observations each, was 0'018° C. 



SECTION IV. 

The Control op the Temperature of the Laboratory. 

22-25. The temperature chosen is dictated by the facility of its maintenance. Four such tempera- 52 

tures are used, namely, 15°, 19*5°, 23° and 26° C. The great majority of observations 
has been made at 19'5° C, which is a suitable temperature for an inhabited room. The 



* From the Greek iw^as, which signifies a body of nine. 



MR J. Y. BUCHANAN ON THE 

PAGE 

maintenance of a, constant temperature in the laboratory requires careful study, details 
of wliich, Avith examples, are given. The room used as laboratory should be of moderate 
dimensions, because it is to be occupied only by the experimenter, who must have 
absolute control over it. It should be illuminated by the light of the northern sky, 
and the direct rays of the sun must be absolutely excluded. When these primary con- 
ditions are given, the experimenter must do the rest. 



SECTION V. 
Tables. 

26. A. General Tables Nos. 1 to 37, giving, the facts of observation. 61 

27. B. Tables Nos. 38 to 61, giving particulars relating to the exactness of the determinations 73 

of the specific gravity given in Tables A, in cases in which two hydrometers were used. 

28. C. Tables Nos. 62 to 71, giving a summary of the specific gravity of the solutions of 80 

different salts at different temperatures. 

29. D. Tables Nos. 72 to 81, giving a summary of the increment of displacement, v, caused 84 

by the dissolution of ?n gram-molecules of salt in 1000 grams of water at different 
temperatures. 

30. E. Tables Nos. 82 to 91, giving the values of v/m, that is, the mean increment of dis- 89 

placement calculated for the dissolution of 1 gram-molecule of salt in 1000 grams of 
water at different temperatures. 

31. A. Tables Nos. 92 to 103, giving the facts of observation for strong solutions. 94 

32. Tables Nos. 104 to 124, of the Classes C, D and E, for strong solutions, 98 



SECTION VI. 

General Description of the Tables. 

33. Explanation of symbols in tables of Class A. With the exception of the determination 100 

of temperature, the result of every series of observations depends only on determina- 
tions of weight, and is independent of the work of others. The necessity for study 
and practice before the experimenter can be confident of his power to control the tem- 
perature of the solution and of the laboratory is insisted on. Failure to appreciate this 
has interfered with the general use of the hydrometric method. 

34. The measure of the displacement of a body having the temperature T is the weight of dis- 101 

tilled water having the temperature T which the body displaces when totally immersed 
in the water. Under this definition the unit of displacement is the space occupied by 
1 gram of water at T. The symbol for the unit of displacement is G^ or G^ , in which G 
represents 1 gram, and T or f the common temperature of the body and of the water 
displaced by it. When the unit of weight is the kilogram, the unit of displacement is 
expressed by the symbol K,p or K(. The eff'ect of concentration on the displacement 
produced by dissolving a given amount of salt in water or in solution is pointed out. 
Dilution of solutions for which m is > 1/8 produces contraction ; when m is less than 
1/16 expansion occurs in many cases. This could not have been ascertained except by 
the liydrometric method. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 5 

PAR. , , PAGE 

35. Discussion of tables of Class B, dealing with the exactness of the results. 103 

36-37. Discussion of tables of Class C, dealing with the specific gravities of the solutions. Dis- 104 

cussion of tables of Class D, dealing with increments of displacement, and the effect on 
the solutions of salts of the double ennead (MR, MRO3) for which m = 1/16 is shown in 
a table and a diagram. The tables of Class E exhibit the comparative volumetric effect 
produced by dissolving different quantities of different salts in 1000 grams of water. 
Each entry in these tables is derived from the corresponding entry, v, in the corre- 
sponding table of Class D by increasing it in the proportion m: 1, whence we obtain 
the value of vjm. 

38. Contains a table giving the specific gravities and the increments of displacement for solu- 105 

tions of all the salts of the double ennead (MR, MRO3) for which «i = l/16. The 
values of the increments of displacement are also exhibited graphically in a diagram. 



SECTION VII. 

The Displacement of the Solutions. 

39. The changes of displacement produced in a constant quantity of water by the dissolution 107 

of successive quantities of a salt in it are compared with those which would take 
place under one of two hypotheses. 

40. First Hypothesis. — It is assumed that, when a quantity of salt, insufficient for saturation, 108 

is dissolved in a quantity of water, it takes possession of the quantity of water which 
it requires in order to produce a saturated solution, and saturates it, after which the 
saturated solution disseminates itself through the remaining water, forming a simple 
mixture with it. If this law is followed by the solutions of a particular salt, then equal 
increments of salt dissolved in a constant quantity of water produce equal increments 
in the displacement of the solutions. This is expressed by the equation 

■ — = Const. 
an 

41. Second Hypothesis. — It is assumed that, when a quantity of salt, insufficient to produce 108 

saturation, is dissolved in a quantity of water, it exercises no selection, but salinifies 
every particle of the water alike, producing a homogeneous solution of uniform concen- 
tration ; and that, when a second quantity of salt, equal to the first, is dissolved in this 
solution, it intensifies its salinity uniformly and produces an increased displacement, 
which bears the same proportion to that of the first solution as the displacement of the 
first solution bore to that of the original quantity of water ; further, that when a third 
equal quantity of salt is added to the solution of the second quantity, it intensifies its 
salinity uniformly and produces an increased displacement, which bears the same relation 
to that of the second solution as the displacement of the second solution bore to that of 
the first, and as that of the first bore to that of the original water ; and so on. Con- 
formity with this law is expressed by the equation 

^i2§^ = Const. 
an 

42. A table for a hypothetical case is given, which affords the means of comparing the effect 110 

produced by diluting or concentrating a given solution with that which would be pro- 
duced if it took place in terms of the first or second of these hypotheses. 

43. When the tables in this memoir are studied, it is found that in the solutions of the majority 111 

of the salts the values of di^jdm and vjm reach a minimum for the values of m in the 



FAR. 



MR J. Y. BUCHANAN ON THE 

PAGE 

vicinity of 1/32, and that they increase whether the concentration is increased or 
diminished. In the ennead MR, the solutions of the caesium salts and the iodides 
approach most nearly to conformity with the law of the first hypothesis. The solutions 
of chlorides and salts of lighter molecular weight conform more nearly to the geometric 
law of the second hypothesis. 

44. When the solutions of a salt follow strictly the law of the second hypothesis, the general 111 

expression for the displacement of a solution containing mMR in 1 kilogram of water 
is Am = Aj"', where A^. expresses the displacement of the solution when m=l. When 
the solution does not follow this law exactly, the displacement for any particular value 
of m is expressed by A„i = Aj''. Then the degree in which the solution conforms to the 
law is indicated by the difference x-m when in is greater than 1. For solutions where 
m is less than 1, and is expressed by vulgar fractions, the expression a;-m is replaced 
by 1/m - Yjx. 

45. The displacements of most of the solutions are treated in this sense in Tables I. to VII. 112 

Tables VIII. to X. give for a number of solutions the values of log A„i/iog A„ = a; ; or, 

~2 

the exponent of the displacement A,„ when the exponent of A is taken as unity. If 
the solutions conform to the geometric law of the second hypothesis, the value of x is 2. 



SECTION VIII. 

Comments on the Chakgbs in the Values of (i A - u for Different Values of m in the 
Case of Solutions op Individual Salts of the Type MR and MRO3. 

46. The increment of displacement {v) due to the dissolution of m gram-molecules of a salt in 1000 116 

grams of water may be looked on as being the result of two operations, namely ; (a) the 
dissolution of m/2 gram-molecules of the salt in 1000 grams of water, which produces 
the first increment of displacement ; and (b) the further dissolution of m/2 of salt in the 
solution so formed, which produces the second increment of displacement. These incre- 
ments of displacement are rarely alike ; the second portion of salt dissolved generally 
produces a greater increment of displacement than the first, and this has been very 
generally held to be the law. One of the principal motives for making this research 
was to find oiit, by the use of the more refined hydrometric method, if there is any point 
in the dilution of a saline solution at which further dilution is accompanied by expansion 
in place of contraction. The general result of the Avork is to show that in the solutions 
having the concentrations here used, where m is less than 1/16, cases of expansion on 
dilution are not uncommon. A table gives the values of m for which the value of 
c? A - w is positive and becomes negative for the next lower value of m. That is, the 
value of cZA - w changes sign at some concentration lower than that indicated by m and 
higher than that indicated by w2/2. 

47. The method of treating the displacements of the solutions of the salts of the enneads 117 

MR and MRO3 is described in the case of solutions of KCl. 

48. The influence of possible error in the determinations of specific gravity on the values of 118 

cZA - u is discussed. 

49. A table furnishes evidence of the agreement in the results obtained by different experi- 119 

menters at different dates and using different instruments. 

50. A specimen table indicates the stages in tlie calculations used in the discussion of the 121 

values of v and rZA in the case of solutions of KCl of different concentrations. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 7 

PAR. PAGE 

51-52. Tables I. to XVIII. illustrate the method of arriving at the volumetric effect produced 123 

by changing the concentration of solutions of each of the salts of the two enneads. 

53. Summary table giving the volumetric effect j^roduced on changing the concentration of 129 

certain solutions of these salts. 

54. Remarks on the tables referring to solutions of the salts MR. 129 

55. Remarks on the tables referring to solutions of the salts MRO3. 131 

SECTION IX. 

Notes on the Values of v for the Enneads MR and MRO3. 

5G. Tables I. and II. give details regarding the change of displacement in solutions of salts of 132 

the ennead MR when changes are effected in the metal or the metalloid of the salt. 

57. Tables III. and IV. give corresponding values for solutions of salts of the ennead MRO3. 134 

Table V. gives corresponding differences between the value of v in solutions of salts of 
the ennead MRO3 when RO3 is replaced by R. 

58. Remarks on the tables relative to solutions of the ennead MR. 135 

59. Remarks on the tables relative to solutions of the ennead MRO3. 137 

60. Consideration of the effect produced on their solutions by the addition of the three oxygen 139 

atoms to the salts of the halides to form the corresponding salts of the oxyhalides. 

61. A general comparison and summary of the variation in the values of the mean increment of 139 

displacement for dilute solutions of salts of the two enneads MR and MRO3. Diagram 
illustrating the variation of v/m with m for values of m= 1/32 to 1/512. 

62. The eighteen salts of the double ennead (MR, MRO3) can be divided into three hexads, 142 

the members of each hexad containing a common metallic element K, Rb or Cs, and 
into three other hexads having a common nietalloidal element CI, Br or I. The relation 
between the values of v/m and m for the three hexads having the nucleus CI, Br or I 
for solutions 1/32 gram-molecule of salt and under are represented graphically in the 
diagram in §61. 

63. Consideration of the order in which the salts of each hexad follow each other when 142 

arranged in ascending order of v/m without paying attention to their numerical values. 
Graphic representation of each hexad of salts by a hexagon, the centre of which is 
occupied by the common element, metal or metalloid, as nucleus. The angles of the 
hexagon are supposed to be occupied by the residues of the salt after the abstraction of 
the common element, arranged in ascending order of magnitude of v/m, the lowest value 
occupying the lowest angle on the paper, and the other values occupying the other 
angles seriatim in ascending order of magnitude and going round from left to right. 
Inside each hexagon we have the common element M or R, and above it the value of ?n 
for the particular concentration. For concentrations higher than m = 1/64 the arrange- 
ment of residues is the same as that given for m = 1/64 in the six hexagons corresponding 
to the common elements CI, Br, I, K, Rb, Cs. The hexagon corresponding to the 
nucleus R and concentration m is represented by the symbol m [R], as, for example, 
1/64 [CI]. The residues of the hexad after the abstraction of the common element CI 
are K, Rb, Cs, KO3, Rb03, CsOg, and these residues, in conformity with the values of 
v/m which correspond to them, follow each other in this order round the corners of the 
hexagon considered. 

64-66, Twelve hexagons of type m[R] and twelve of type ?;i[lVI] are given in the text, and the 144 

different ordinal sequences of the residues for different hexads as exhibited in the 
hexagons are fully discussed. 



MR J. Y. BUCHANAN ON THE 



SECTION X. 



Experimental Observations on the Displacement op Solutions of Sodium Chloride. 
PAR. page 

67. Although sodium chloride is not a member of the enneads which form the principal material 148 

of this research, its importance in nature justifies its inclusion in it. General table of 
results of experiments made on solutions of sodium chloride varying in concentration 
from 1 gram-molecule to 1/512 gram-molecule per 1000 grams of water. 

68. Preparation of solutions forming arithmetic series. 149 

69. Tables giving results of solutions for which m forms a geometric series of the usual type. 150 

70. Discussion of specific gravity. 150 

71. Discussion of displacement. 150 

72. Experiments on solutions forming arithmetic series. 151 

73. Series of experiments on solutions having the common difference dm = 1/128. 151 

74. Discussion of differences of displacement. Table confirming the reality of the remarkable 151 

changes of displacement in high dilutions. 

75. Table giving further confirmation of this. 152 

76. Table giving series of experiments on solutions having the common difference of concentra- 153 

tion dm = 1/64. 

77. Discussion of displacement and differences of displacement. 153 

78. Diagram giving graphic representation of remarkable changes in the values of vjm at high 154 

dilution. 



SECTION XI. 
The Principle and Construction of the Open Hydrometer. 

79. The hydrometer is left open so that its weight may be increased or reduced by additions 155 

to, or removals from, its ballast or internal load. The extent to which additions can 
be made to the weight supported by the stem of the instrument is limited by its stability. 
The Challenger hydrometer, which was closed, could be used in a saturated solution 
of sodium chloride carrying an accessory weight of 32 grams. In a solution of greater 
density, the external weight required produced a "list." 

80. The open hydrometer may be of the same size as the closed instrument, and is made after 155 

the ordinary pattern, but the millimetre scale is etched on the stem, the paper scale 
being impossible when the internal load is to be varied. When concentrated solutions 
of salts which are both very soluble and very expensive are used, it is convenient to 
use a hydrometer of less bulk than that of the closed instrument. 

81-84. The weight of the instrument is not constant, because the internal ballast is altered from 157 

time to time, and the weight of the air in the hydrometer varies with the meteorological 
conditions. The first step in the preparation of the hydrometer for use is to weigh the 
glass instrument empty as it comes from the glass-blower. It is then weighed with 
the ballast (lead shot) in it. Knowing the density of the glass and that of the lead, 
we obtain the volumes of the glass and the lead respectively. When the instrument is 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 

R. 

immersed in distilled water and accessory weights are added until it floats almost totally 
immersed in the water, the sum of the accessory weights, the weight of the glass, and 
that of the lead give the weight of the water displaced, and from this the external 
volume, or the displacement, of the instrument is obtained. The volume of the air 
inside the hydrometer is then given by the difference between the external volume of 
the instrument and the sum of the volumes of the glass and lead. Full details of the 
experimental work required in order to obtain the exact weight of the hydrometer 
under particular conditions are given. 



SECTION XIL 
NuMEBiCAL Details illustrating the Use of the Open Hydrometer. 

85. Details are furnished relating to the experimental determination of the weight of the 165 

liquid displaced by the instrument designated Hydrometer A. 

86. Scheme for logging the observations made with the hydrometer in the experimental 167 

liquid at the selected standard temperature, T. A table gives numerical examples using 
(a) distilled water, and (b) a 7 gram-molecule solution of rubidium chloride. 

87. Correction for the buoyancy of the non-immersed portion of the stem. As in the case of 169 

the closed hydrometer, this is practically inappreciable. 

88. The degree of accuracy attainable l)y the use of the open hydrometer is illustrated by the 169 

results of five series of observations, each series consisting of eleven independent 
observations, made in a solution of calcium chloride containing 6*3 gram-molecules of 
CaCig in 1000 grams of water. Of the five series, three were made vvith hydrometer A 
and two with hydrometer B. The values of the mean specific gravities, S, furnished 
by each series and its probable error ( ± r^), expressed in units of the sixth decimal place, 
are collected in a table. 

89. Precautions to be taken in order to secure trustworthy results. My practice in the 171 

CJiallenger was, when I began hydrometric observations, to lock the door, and I still 
adhere to this practice. Attention is called to the effect of the low specific heat of 
concentrated solutions, such as 6 3 CaClg, in increasing the thermal nimhleness of the 
solution. 

.90. Table of specific gravities calculated from single observations made with hydrometers A 172 

and B when floating in a solution of calcium chloride containing 6"3 gram-molecules of 
CaClg in 1000 grams of water. This table includes the individual observations the 
means of which were given in § 88. 



SECTION xm. 

On the Specipic Gravity and Displacement of Solutions op Salts of the Ennead MR which 

HAVE nearly the SAME MoLECULAR WEIGHT AND MAY BE LOOKED ON AS " ISOMERIC." 

91. These salts are RbCl, KBr,K 51+1. 172 

92. A table gives the results of specific gravity determinations made upon the solutions of 173 

these salts at 19 "5' C. 

93. The solubility of each of the above salts was determined. 174- 
TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 2 



10 MR J. Y. BUCHANAN ON THE 

PAR. PAGE 

94. The specific gravities of the solutions were adju'^ted to the value which they would have if 174 

their gram-molecules had the uniform weight 121, which is the actual molecular weight 
of tlie heaviest of the three, namely, rubidium chloride. This adjustment of the 
molecular weights does not materially affect the relations of the solutions as regards 
their specific gravities. The displacements of these solutions are compared. 

95. The differences of displacement are discussed. 175 

96. The displacement of solutions of the artificial isomer (K ) is considered in reference 176 

to the displacement of solutions of the constituent salts. 

97. Comparison of displacement of solutions of the artificial isomer as obtained by experiment 177 

and as calculated. 



SECTION XIV. 

The Specific Gravity and the Displacement of Solutions of the Chlorides of 
Beryllium, Magnesium and Calcium. 

98. Preparations of weak solutions — m = 1/2 to 1/1024 — of each of the three salts, and of strong 178 

solutions of MgCl2 and CaCla for 7n = 1 up to supersaturation. The solutions saturated 
with MgClj and CaClj at 195° C. contain 5'918 and 6'613 gram-molecules respectively 
in 1000 grams of water. A supersaturated solution of magnesium chloride contained 
5 982 MgClg in 1000 grams of water. This solution was formed with moderate absorp- 
tion of heat and crystallised very readily. The supersaturated solution of calcium 
chloride contained 7 '225 CaClj per 1000 grams of water. This solution was formed 
with great absorption of heat, and offered considerable resistance to crystallisation. It 
was found that when the quantities of the crystallised salt MgCl26H20 and water used 
were such as to produce a solution containing about 2 MgClj per 1000 grams of water, 
there was an appreciable liberation of heat. When further salt was dissolved this gave 
place to absorption of heat, and, at saturation, the temperature of the solution was lower 
than the initial temperature of the water used. 

99. Table giving the results of specific gravity determinations made upon solutions of 179 

the chlorides of beryllium, magnesium and calcium of different concentrations at 
19-5° C. 

100. While the bases BeO, MgO and CaO give an alkaline reaction with litmus paper, the 180 

chlorides of magnesium and calcium are neutral, while that of beryllium is acid. The 
beryllium chloride solution was made from the sulphate by double decomposition with 
barium cliloride. The more dilute solutions were prepared by diluting the more con- 
centrated ones. The specific gravities of the strong solutions were made with open 
hydrometers A and B, and those of the weak solutions with closed hydrometers Nos. 3 
and 17. Comparison of (S - 1) with m. A table is given from which it is apparent that 
the values of (S - 1) produced by dissolving 1/2 MR in 1000 grams of water are exactly 
proportional to the molecular weights of the salts in the case of MgClj and CaClj, 
and that this proportionality is maintained for values of m = l/16 and 1/128. In 
the case of beryllium chloride the proportionality fails. The specific gravities of 
the solutions of beryllium chloride for which m = 1/512 and to= 1/1024 fall below 
unity, from which it follows that the displacement of these two solutions must be 
greater than the sum of the displacements of the salt and water which they 
respectively contain. The values of dS for solutions of CaClj which are near 
saturation are discussed. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 11 

PAR. PAGE 

101. Discussion of the values of d^/dm and v/m especially for solutions for which m is less 182 

than 1. 

102. The values of dA-vave discussed. The principal feature of the table illustrating them 183 

is the pronounced expansion which accompanies the dilution of solutions of beryllium 
chloride for which m is less than 1/16. 

103. The relations between the exponents of the solutions x and m are discussed, and a table 184 

of the values of log A„,/log A,„ for solutions of each of the three salts is given. 



SECTION XV. 

On a Remarkable State op Unrest in a Supersaturated Solution of 
Calcium Chloride before Crystallising. 

104. The supersaturated solution was 7 '225 CaCl2 + 1000 grams of water. It was expected that 185 

this solution would crystallise easily and furnish a truly saturated solution. As it 
showed no inclination to crystallise although every opportunity was offered it to do so, 
it was adopted as an example of a supersaturated solution peculiarly adapted to closer 
study. Table I. contains the constants of the open hydrometers A and B, as loaded for 
the experiments of this section, (a) when floating in distilled water, and (6) when float- 
ing in the supersaturated solution of calcium chloride. 

105. The experiments showing the state of unrest were made 11th May 1910, in the Davy- 186 

Faraday Laboratory. A series of observations had been made with each hydrometer, 
and further observations were proceeding when it was noticed that discrepancies between 
successive readings and corresponding ones in the earlier experiments made with the 
same added weights were occurring, and that these were far greater than any which 
could be attributed to error of observation. They persisted while four series of obser- 
vations were made, two sets with each hydrometer, and were so great that in the fifth 
series of observations it was necessary to reduce the initial added weight in order that 
the complete series of observations might be made. The temperature of the solution 
was perfectly constant at 19"5° C. during each series. 

106. After removal of the hydrometer from the experimental solution on completion of the fifth 187 

series of observations, the solution was stirred as usual with the standard thermometer, 
and its temperature was found to be 19'50°, that of the air being 19"30°. It was not 
until after these observations had been made that a cloudiness indicating the commence- 
ment of crystallisation appeared in the solution. It increased rapidly, and the tem- 
perature rose smartly to 23' 16° C, and remained constantly at that temperature from 
1.10 p.m. to 2.35 p.m., a period of 85 minutes, when tlie temperature began to fall. 
The supersaturated solution (7'225 CaClg-l-lOOO grams of water) contained 44'48 per 
cent. CaClg. When the temperature of the mixture of crystals and solution had 
fallen someAvhat, the cylinder was placed in water having the temperature 19'3°, and 
was cooled to 19'5°. The mother-liquor was then found to have the specific gravity 
1-423500, and to contain 42-33 per cent, of CaClj. 

107. The crystals along with the mother-liquor were then heated in the cylinder to a tempera- 187 

ture of 30° C. by placing the cylinder in a water-balb of about that temperature, and 
keeping it there until tlie crystals were redissolved. The system was then allowed to 
cool in the air, the temperature of which remained constant at 19-3°, and the tempera- 
ture of the cooling liquid was taken at intervals of 30 seconds. The series of observa- 
tions extended over 41 minutes, during which the temperature fell from 23-82° to 



12 MR J. Y. BUCHANAN ON THE 

PAR, PAGE 

21 '99°, and the solution remained liquid to the end. The cooling had proceeded 
for 13 minutes before the temperature fell to 23"16°, and the loss of heat was 
taking place quite regularly. -The following are the temperatures observed at each 
^ minute for 2 minutes before and 2 minutes after the temperature of 23"16° was 
passed : — 

Time in minutes : -2-0 -1-5 -TO -0-5 00 +05 I'O 1-5 20 
Temperature: 23-23° 23-21° 2319° 23-17° 23-16° 23-14° 23-12° 23-09* 2307° 

During the 4 minutes the temperature fell 0-16°, whence 0-04° per minute represents 
the mean rate of fall of temperature when the system has the temperature 23-16° and 
is cooling in air of constant temperature 19-30° C. 

108. Calculation of Heat liberated during Crystallisation. When crystallisation was started, 188 

the temperature of the system rose in less than a minute from 19-5° to 23 '16°. During 
this phase the temperature of the cylinder with its contents was raised 3-66°. The 
heat liberated in this act was found to be 2217 gram-degrees (gr.° C). During the 
second phase the rate of liberation of heat was equal to its rate of dissipation, which 
was represented by a fall of temperature of 0-04° per minute. This was maintained 
for 85 minutes, which requires a liberation of 2059 gr.° C. of heat; so that the total 
heat liberated in the act of crystallisation was 4276 gr.° C. 

109. Verification of the constitution of the crystals as CaCl26H20. According to Thomsen, the 188 

heat of solution of CaCl26H20 is —4340 gr.°C. ; therefore on thermal evidence alone 
215-5 grams or 0-984 CaCl._,6H20 has separated out. On the basis of analytical estima- 
tions made on the supersaturated solution and the mother-liquor, 210-3 grams or 
0-96 CaCl26H20 must have sei^arated out. The agreement of these two computed 
values is excellent. 

110. Descri[ition of Tables 11a. and IIb. In Table IIa. are given the individual observations 189 

of specific gravity forming together the five series, of eleven observations each, on tlie 
supersaturated solution 7-225 CaCl2 when it was exhibiting the state of unrest which 
preceded crystallisation. In Table IIb. are given the individual observations forming 
five series of eleven observations each, on the non-saturated solution 6 '3 CaClj. 
Table lie. contains the imlividual observations forming three series of eleven observa- 
tions each, on the super.-atiirated solution 7-196 CaClg- It crystallised suddenly after 
the third series. Table III. forms a time-table of the observations on the supersaturated 
solution 7-225 CaCl2. 

111. Discussion of conditions of temperature maintained while the operations recorded in 193 

Tables IIa. and IIb. were being made. 

112. Further discussion of Tables IIa. and IIb. Considering the five mean specific gravities of 194 

7-225 CaClg, it is found that the maximum amplitude of variation is 689 units in the 
sixth decimal place, while the five mean specific gravities of 6-3 CaCl2 exhibit a 
maximum amplitude of only 26 such units. When the observations of individual 
series are considered, the maximum amplitude of variation in the fifih series for 
7-225 CaCl2 is 833 in the sixth decimal place. These large and rapid variations of 
specific gravity in the supersaturated solution furnish the evidence of the state of 
unrest existing in it. 

113. The displacement of the solution is subject to variations corresponding to those of the 195 

specific gravity. Tiiey afford evidence of spasmodic acts of expansion and contraction, 
not accompanied by any change of temperature of the solution or of the external 
pressure to which it is subjected. They exhibit a veritable species of labour going on 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 13 

PAR. PAGE 

in the solution in its efforts to become a mother-liquor. In this it is finally successful, 
but not before it has succeeded in forcing the door which confined its store of heat. 
The birtli of the crystal was synchronous with and dependent on the liberation 
of heat. 

114. It is shown that the change of displacement which occurred in the transition of the solution 196 

7*225 CaCl2 from a condition of supersaturation to that of a mixture of saturated solu- 
tion and crystals at the common temperature 19"5° C. is a shrinkage amounting to 2-2 
per cent, of the original volume of the supersaturated solution. 

115. Comparison of the behaviour of supersaturated solutions of MgClg and CaClj with respect 196 

to readiness in starting, and heat exchange accompanying, crystallisation. The variations 
of the density of the liquid before the first element of crystal appears revealed only by 
the skilled use of the hydrometer. The diagram illustrates the clianges of displacement 
corresponding to the changes of density in the 7"225 CaClg and the 7"196 CaClj super- 
saturated solutions, compared with the accidental changes observed in tlie stable solu- 
tion 6'3 CaCl2. In the case of tlie 7'225 CaCl2 solution the state of unrest persisted 
during the 140 minutes that the experiments lasted, and it seems to be not improbable 
that a supersaturated solution is never at rest even in a closed vessel. 

116. Analogy between the crystallisation of a supersaturated saline solution and the formation 197 

of ice when a non-saturated solution or when pure water is cooled below its freezing 
point. When the mass of water is small and the capucity for heat of the vessel which 
contains it is large, the temperature of the system may be reduced so far that when 
freezing begins the whole of the water may be frozen without the temperature of the 
system rising to 0° C. Experimental illustration of this. Possibility of detecting 
oscillations of density in water before freezing begins, by determining its specific 
gravity hydromelrically with the necessary precautions in a room having a constant 
temperature between - 4° and - 5° C. 

117. Calculation of the increment of pressure required to countoract the stretching of the 200 

7'225 CaClg solution before the beginning of crystallisation. It is found to be 38 
atmospheres. 

118. Resemblance between the state of unrest preceding the crystallisation of a supersaturated 200 

solution and that preceding the liquefaction of a gas under a pressure not inferior 
to its critical pressure, when its temperature is reduced slightly below its critical 
temperature. 

119. It is only in the conditions of Andrews' experiment on COg that we can witness a substance 201 

persisting in the gaseous state under a pressure greater than its critical pressure, and 
having a temperature lower than its critical temperature, because it is only when the 
gas and the envelope which contains it have been maintained at a temperature higher 
than the critical temperature of the gas, that the inner walls of the envelope have a 
chance of being perfectly dry, that is, free from every trace of the liquid substance. 
We do not know the temperature at which a dry gas can liquefy on the dry walls of its 
envelope, but so soon as the first, even the minutest, trace of the liquid substance 
appears, the temperature of liquefaction is defined, because the gas is then condensing 
on itself as a liquid. 



14 MR J. Y. BUCHANAN ON THE 



SECTION XVI. 

The Determination op the Specific Gravity of the Crystals op a Soluble Salt by Displacement 
IK its own Mothbb-Liquor, and the Volumetric Relations between the Crystals and the 
Mother-Liquor which are established by the Experiment. 

PAR. PAGE 

120. This work was undertaken owing to the arrival of the great anticyclone or heat-wave of 202 

the summer of 1904, which made observations of specific gravity at 19 '5° impossible. 
The liquid in which every soluble salt is quite insoluble is its own mother-liquor at the 
temperature at M'hich the one parted from the other. It was in this liquid that the 
specific gravity of the crystals of tlie salts of the two enneads MR and MRO3 was 
determined. It is obvious that this method is applicable only to salts which have a 
mother-liquor, such as KCl, RbBr, CaCIgGHgO, BaCl22H20. It is inapplicable to salts 
such as CaCly, BaClg, and the like, which have no legitimate mother-liquor. The anti- 
cyclone prevailed throughout the greater part of July and August 1904, during which 
time the determinations of the specific gravity of the crystals and the mother-liquors 
of the salts of the ennead MR were determined. 

121. Precautions to be observed in making the experiment. 203 

122. Determinations of the solubility of the salts RbBr, Rbl, CsCl, CsBr and Csl were made, 203 

as there were no published data regarding them. The preliminary experiments are here 
described. 

123. Contains Table I. in which the experimental data and details are given in full in the case 204 

of one salt, namely, CsCl. All the weights as given represent the weight in vacuo. 
Further necessary details of the experimental method are here given. 

124. Precautions to be observed when bringing the crystals together with the mother-liquor in 206 

the pyknometer. The experimenter must realise that their common temperature when 
mixed is to be exactly that of crystallisation or equilibrium, and he must take such 
measures as his experience dictates to arrive at this end. 

125. Contains Table II., which gives for each salt the temperature, T, of equilibrium between 207 

crystals and mother-liquor, and in condensed form the experimental data of the determina- 
tion of S, the specific gravity at T of the mother-liquor, that of water at the same 
temperature being unity ; of m, the concentration of the mother-liquor in gram-mole- 
cules of salt per 1000 grams of water; and of D^, Dj jDg, the three observed values, as 
well as D, the finally accepted value of the specific gravity of the salt, all at T, and 
referred to that of water at the same temperature as unity. 

126. General discussion of the results. 209 

127. Contains Table III., giving numerical relations between the crystallised salts of the ennead 209 

MR and their mother-liquors. 

128. Discussion relative to the mother-liquor. 211 

129. Consideration of saturated solutions as products of substitution. 212 

130. Comparison of the displacement of the salt in crystal and the increment of displacement 213 

of 1000 grams of water which is produced by its dissolution. It is shown that the 
crystallisation of the potassium and rubidium salts of the ennead must be hindered 
by increase of pressure, while that of the caesium salts must be helped by the same 
agency. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 15 

PAR. PAGE 

13 L Account of similar experimental researches for the crystals and mother-liquors of the salts 214 

of tlie ennead MRO3. The investigation was made on a plan exactly similar to that 
used in the case of the salts of the ennead MR. Table IV. corresponds to Table IT. of 
the ennead MR, and gives in a condensed form the data bearing upon the observed 
values of the specific gravity of the salts. 

132. Table V. gives the results of observations made with the crystals and mother liquors of 214 

the salts of the ennead MRO3. It is arranged on the same plan as Table III. for the 
salts of the ennead MR, and consists of a number of sub-tables, the nature of each of 
which is specified in its title. 

133. Contains a table giving the specific gravities, D, of the salts of the ennead MRO3, and 215 

their differences. The observations recorded in Tables IV. and V. are further 
discussed. 

134. The molecular displacement, MRO3/D, of the crystal expressed in grams and molecules of 218 

water is considered. 

135. The molecular concentration of the mother-liquor is discussed. The value of m does not 218 

in any case exceed 1/2. The values of the concentrations are derived from the specific 
gravity of the mother-liquor. 

136. The values of -:- 3 -— are discussed. As they are all positive, crystallisation is in 218 

every case accompanied by expansion. 

137. In a table are given the differences between the molecular displacements in crystal of 218 

the corresponding salts of the two enneads, MRO3 and MR, and these are 
commented on. 

138. Concluding remarks. 219 

Appendix A. — Densities of the solutions at T. 

Appendix B. — Table giving the number of series as well as the number of single 
observations made with the various Hydrometers, from which the results recorded in 
this Memoir were obtained. 

Index, 226 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 17 



Section I. — Introduction. 

§ 1. The Principles of Archimedes* — It is well known that the mechanics of floating 
bodies, and the laws which govern their equilibrium, were established and enunciated 
by Archimedes, the Sicilian, in the third century before our era. The following pro- 
positions, demonstrated in the first book of his treatise on this subject, embody the 
fundamental principles of the hydrometer : — 

(a) The surface of any fluid at rest is the surface of a sphere whose centre is the 
same as that of the earth. 

(h) Of solids, those which, size for size, are of equal weight with a fluid will, if 
let down into the fluid, be immersed so that they do not project above the surface, 
but do not sink lower. 

(c) A solid lighter than a fluid will, if immersed in it, not be completely submerged, 
but part of it will project above the surface. 

{d) A solid lighter than a fluid will, if placed in the fluid, be so far immersed that 
the weight of the solid will be equal to the weight of the fluid displaced. 

((?) If a solid lighter than a fluid be forcibly immersed in it, the solid will be 
driven upwards by a force equal to the difl"erence between its weight and the weight 
of the fluid displaced. 

{/) A solid heavier than a fluid will, if placed in it, descend to the bottom of the 
fluid, and the solid will, when weighed in the fluid, be lighter than its true weight 
by the weight of the fluid displaced. 

Archimedes considered only one solid and one fluid, and his laws regulate exactly 
what takes place in such a system when the solid is totally immersed in the fluid ; 
or, if only partially immersed in it, when the non-immersed portion of the solid is 
immersed in no other fluid — in other words, when the experiment is being made in 
a vacuum, or in a medium the density of which is insensible. When, however, the 
experiment is being made in air, it is not necessary to postulate that its density 
is insensible ; Archimedes' laws still hold good, only the solid falls to be considered 
as divided into two, one of which is completely immersed in the one fluid (the 
liquid), and the other is completely immersed in the other fluid (the air). If the 
solid was immersed in three fluids, as, for instance, water, oil, and air, and floated 
at rest when part of it was immersed in each of these fluids, it would fall to be 
divided into three portions, each of which is totally immersed in one of the three 
fluids, Archimedes' laws would still be applicable, and the final total effect would 
be the sum of the partial effects. 

§ 2. Hydrometer suitable for Demonstrations on the Lecture Table. — I constructed 
an instrument of this kind for use in lectures which I gave as assistant in the 

* The IVorhs of Archimedes, editedin Modern Notation, by T. L. Heath, Sc.D., Cambridge University Press, 1897, 
pp. 253-268. 

TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 3 



18 



MR J. Y. BUCHANAN ON THE 



M 



m 



- 5 



University Laboratory in Edinburgh, under Professor Crum Brown, in the years 
1869 to 1872. A description of it was published in the Berichte der Deutsclien 
Chemische Gesellscha/t (1871), iv. 338. Fig. 1 is a sketch of it. The stem of 

this instrument was made of glass tube having an external 
diameter of about 1 centimetre, and a truly circular section of 
uniform diameter. A slip of paper is attached inside the stem. 
It is graduated on any convenient scale of equal lengths, and 
the divisions are numbered upwards and downwards from the 
zero point in the middle. The numerals from upwards have 
the positive sign, and those running from the downwards have 
the negative sign. 

The hydrometer is ballasted with mercury or shot, so that, 
in its completed state, it sinks in the liquid used, at the atmo- 
spheric temperature, exactly to the zero division in the middle 
of the stem. 

The lower extremity of the instrument takes the form shown 
in the figure, terminating in a hook K. The upper extremity 
of the stem is closed with a cork, to which a suitable disc of 
cardboard, M, is attached by sealing-wax. 

The hydrometer was originally constructed in order to illus- 
trate the determination of the specific gravity of solid bodies. 
The liquid in which it is to be immersed may be distilled water, 
but other liquids, for instance sea-water, may also be used. 
It is contained in a suitable cylinder, and should have the 
temperature of the room in which the experiment is being 
made. 

When it is proposed to exhibit the determination of the 
specific gravity of any particular solid body, the hydrometer 
is immersed in the liquid, in which it sinks until the zero 
division on the scale is exactly in the plane of the surface of 
the liquid. A suitable fragment or piece of the solid body 
is then placed on the platform M, and the extent to which 
the immersion of the stem in the water is increased is noted. 
The solid body is then removed from the platform M and 
attached to the hook K, and the hydrometer is again im- 
mersed in the water. When equilibrium of flotation has been 
established, the immersion of the stem is again read on the scale. 

Let the former of these two numbers be expressed by a and the latter by J), a and 
h are lengths of a cylinder of uniform diairieter and of circular section ; therefore the 
volumes of these cylinders are proportional to their lengths ; and, as the same liquid is 
displaced in each case, the weights of the liquids so displaced are also proportional to 




Fig. 1. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 19 

the lengths a and b. It follows, therefore, that the expression (a — b) represents the 
weight of a volume of the liquid equal to that of the solid body, and that the specific 
gravity of the solid body, referred to that of the liquid as unity, is 

D = 



a-b' 



When a solid body is placed on the platform M, the hydrometer always sinks 
deeper in the liquid ; therefore a is always positive. When a solid body consists of a 
substance which is denser than the liquid, then, when it is attached to the hook K and 
is immersed with the instrument in the liquid, it causes the hydrometer to sink deeper 
in it, and b is also positive in this case. When the substance of the solid body is less 
dense than the liquid, a is positive as before ; but when the body is attached to the 
hook, and is immersed with the hydrometer in the liquid, it exerts a pressure wpivards, 
which causes the hydrometer to emerge and expose a part of the stem below the point 
0. On this part of the scale the numerals have the negative sign, and the weight of 
the volume of liquid displaced by the solid body is, as before, (a — b) ; and its specific 
gravity is, also as before, 

a-b 

The identity of the expressions of the experimental data in determining the specific 
gravity of substances so dissimilar as, for instance, a stone and a cork never failed to 
arrest the attention of the students. 

It will be noticed that, by using this method, the specific gravity of a solid body 
is obtained ivithout any deter^nination of weight having been made, either' in the 
production of the instrument or in its use, and that the only measurements made are 
those of length. 

§ 3. Usefulness of the Hydrometer in the Study of Mineral Waters. — But the 
hydrometer and its uses had always had a fascination for me. I began to pay particular 
attention to the subject in Wiesbaden, when working as a student with Fresenius, and 
afterwards (1866-67) as an assistant in his private analytical laboratory. During this 
period I became much interested in the mineral waters which abound in the (then) 
Duchy of Nassau and the neighbouring parts of the Rhineland, and especially in the 
Kochbrunnen of Wiesbaden, perhaps the most celebrated of them all. I had great 
curiosity to investigate the variations, if any, in its concentration at different times and 
seasons ; but, as a student, I had to follow the plan of instruction laid down, and, in the 
private laboratory of the final referee in Germany regarding all matters of dispute or 
arrangement which could be decided by chemical analysis, the important and responsible 
work entrusted to me made it impossible for me to occupy myself with anything 
else at the same time. During the voyage of the Challenger, I many times made 
up my mind, on my return to Europe, to visit Wiesbaden and use the hydrometer 
in carrying out a systematic investigation in this sense ; but my intention has not 
been realised. 



20 MR J. Y. BUCHANAN ON THE 

§ 4. The Hydrometer in the ^^ Challenger'' Expedition. — The dispatch of the 
Challenger Expedition was decided before the end of the year 1871, and Sir Wyville 
Thomson, who was then Professor of Natural History in the University of Edinburgh, 
was chosen for its leader. He did me a great honour and a very substantial service in 
selecting me for the post of chemist and physicist of the expedition quite a year before 
the date fixed for its departure. I cannot adequately express the gratitude which 
I feel for the confidence which he thus showed in me, and for the privilege which 
it gave me of taking an active part in this memorable expedition. The expedition 
lasted less than four years, yet these years are fuller of recollections than all the rest 
of my life. 

During the year which elapsed between my selection and my official appointment, 
I occupied myself almost exclusively in preparing for my work at sea, and I considered 
that the specific gravity of the water of the ocean, and its variations, would be one of the 
most important matters for continuous observation. Here I had in view the variations 
of specific gravity which occur in the open ocean and far from all influence of the land. 
These were only imperfectly known, but there was reason to conclude that they were 
confined within narrow limits. 

I chose the hydrometer, or " araeometer," as it is called abroad, because it appeared 
to me to be the only type of instrument which furnished directly the information 
demanded, namely, the specific gravity of the water, and that with the exactness 
required when the variations of specific gravity are so small. 

Even at that early date indirect methods of all kinds were recommended to 
me. In theory, any physical constant of a saline solution, the expression of 
which includes a term depending on its specific gravity, can be used for this 
purpose. But indirect methods «re, m the nature of things, affected with at 
least a double quantity of eirors. There are the errors ivith which the datum 
directly supplied by the vicarious method used is affected, and there are those which 
affect the operation of comjja^'ison by which that datum obtains its densimetric 
interpretation. 

I had then, and I have still, an instinctive dislike of all indirect methods in science ; 
I therefore adhered to my own purpose, believing that, if nothing but manipulative 
difficulties stood in the way, they could be overcome by perseverance and a determina- 
tion not to accept defeat too readily ; and, as is so often the case, the difficulties 
apprehended turned out to be in no way formidable. 

§ 5. In designing the hydrometer I decided that, in the values of the specific gravity 
obtained with it, units in the fourth place of decimals must be exact, and that the 
exactness should be pushed as far as possible into the fifth place. As a knowledge of 
the physical constants of the instrument is of the first importance, I rejected the plan 
of having a series of liydrometers, each to be used in the waters the specific gravity of 
which corresponded to the limits of its scale. I deeided to have one hydrometer, made 
of glass, in which the dimensions of the stem and of the body should be in such pro- 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 21 

portion as to ensure the degree of accuracy above indicated, and provision for extending 
its range of usefulness to sea-waters of all specific gravities should be made by suitable 
alterations of its weight. Considerations of stability suggested attaching the accessory 
weights necessary for this purpose to the lower extremity of the hydrometer. But this 
would involve their being immersed in the liquid the specific gravity of which was to 
be determined, and was therefore inadmissible. The only alternative was to attach them 
to the upper extremity of the stem. A length of 10 centimetres of the stem was 
graduated into millimetres, and the external diameter of the stem was such that its 
graduated portion displaced rather less than 1 gram of distilled water. The body of 
the instrument was constructed so as to have a volume of approximately 160 cubic 
centimetres. The hydrometer was ballasted with mercury, so as to fioat in distilled 
water of ordinary temperature with the whole of the graduated part of the stem exposed. 
The system of accessory weights designed for increasing the range of the hydrometer 
included, as first weight, a small brass table which fitted on to the top of the stem. Its 
weight was designed so that, if the hydrometer alone floated at the lowest division of 
the scale in a particular water, and the table was then affixed to the top of the stem, 
the hydrometer would sink until it floated at a division near the top of the scale in the 
same water. Of the further weights, the first of the series was a mass of brass of about 
the same weight as the table. When the hydrometer carrying the table on the stem 
floated at the lowest division of the scale in a particular water, then, by placing the 
further weight on the table, the hydrometer sank until it floated at a division 
near the top of the scale. The weight of the next weight of the series was made 
approximately double that of the table, so that when the hydrometer, loaded with 
the table and the previous weight, floated at the lowest division on the scale in a 
particular water, and the previous weight was replaced on the table by the present 
one, the hydrometer floated in the same water at a division near the top of the 
scale ; and so on. 

S 6. The series of weights was carried so far that waters of all densities from that of 
distilled water to that of water more dense than that of the Red Sea could be determined 
witli the same hydrometer. 

The accessory weights form roughly an arithmetical series, the common diff"erence 
of which is equal to the first term, namely, that of the little table to be placed on the 
top of the stem and to carry the other weights, when required. As produced, the 
weights fulfilled all the conditions demanded of them, and all that it was necessary to 
know was the exact weight of each, and this was determined. 

From the design of the system of accessory weights, it will be seen that provision 
was made for single observations of specific gravity. Duplicate observations were 
possible only in cases where the salinity and temperature of the water combined to 
produce such a specific gravity that it could be observed with one of the sets of weights 
near the lowest division of the scale. In that case its specific gravity could be obtained 
also with the next higher weight at a division near the top of the scale, because the 



22 MR J. Y. BUCHANAN ON THE 

difference between the successive weights was rather less than that required to immerse 
the divided portion of the stem. 

On rare occasions multiple observations were made on a single water, using ordinary 
decigram weights, but this was found to be very inconvenient. Nevertheless, the 
advantage of multiple observations was clearly perceived, and provision for their being 
made was included in the specification of all later instruments. Also, the system of 
numbering the centimetres on the stem was altered. In the Challenger instrument 
the number 10 marks the lowest division on the stem, and marks the highest. In 
all later instruments the lowest division is 0, and the centimetres are numbered 
1, 2, 3, . . . 10 upwards. 

In every determination of the specific gravity of a sample of water, the weight 
of the volume of it which was displaced by the hydrometer floating in it at an 
observed division on the scale was represented by the sum of the weights of the 
hydrometer, the table, and the accessory weight used. 

The volume of the water so displaced by the hydrometer was arrived at as 
the result of an extensive series of observations made with it in distilled water 
at different temperatures. 

The relation between the weight and the volume of a mass of distilled water 
at all ordinary temperatures, as determined by Kopp, was accepted as correct, and 
was used in reducing the observations made with the hydrometer in distilled water 
so as to arrive at the volume of its body, that is, the whole of the hydrometer 
below the lowest division on the stem at all ordinary temperatures. Its rate of 
thermal dilatability was taken to be constant within the limits of temperature 
considered, and its probable value was obtained by taking the mean of all those 
observed. 

The final result was stated by giving the volume of the body of the instrument 
up to the lowest division on the stem at 0° C. as V, and the rate of its dilatability, 
dV/dt, as e. 

Thus the full specification of the hydrometer, that is, the glass instrument alone, 
is furnished by four data. 

For the hydrometer used in the Challenger they are : — 

Weight Mi vacMo of the hydrometer . . W 160 '2 128 grams. 
Volume of body of hydrometer up to lowest 

division on stem at 0' C. . . V 160-277 c.c. 

Rate of expansion of body per ° C. . . e 0*00455 c.c. 

Total volume of divided stem (100 mm.) . v 0*8650 c.c. 

The specification of the set of accessory weights which were used with this 
hydrometer is as follows : — 



No. . . . 
Weight in grams 



0. 


I. 


II. 


III. 


IV. 


V. 


VI. 


0-8360 


0-8560 


1-6010 


2-4225 


3-2145 


4-0710 


4-8245 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 23 

Weight No. is the small brass table which can be affixed to the top of the 
stem, and on it any further weight that might be required was placed. The distinctive 
number of the hydrometer which was used for all the determinations made in the 
Challenger was 0. In the tabulated results the combination used is indicated in 
the column headed " Number of Hydrometer." Thus, OOV means that Hydro- 
meter No. 0, table No. 0, and weight No. V. were used. The combinations almost 
exclusively used were OOlV and OOV, which weighed 1642633 and 165H98 grams 
respectively. 

In this memoir we make no use of the volume of the hydrometer, because in 
all the experiments the temperature is a constant, and we obtain directly the dis- 
placement, that is, the weight of distilled water displaced by the same volume of 
saline solution, both being at the same temperature, from which we obtain directly 
the specific gravity of the solution at that temperature, referred to that of distilled 
water at the same temperature as unity. This result is arrived at from the two 
observations alone, and is independent of the work of others. 

As it was certain that during the voyage of the Challenger the specific gravity 
of the sea-water would have to be observed at many diff"erent temperatures, it was 
convenient, after having determined the displacement of the hydrometer in distilled 
water at diff"erent temperatures, to express the result in terms of the volume of the 
displacing hydrometer in standard cubic centimetres, but the difference from the 
later practice is only in the form of expression. 

§ 7. In order to obtain all the precision of which the hydrometric method is 
capable, the temperature of the water must remain perfectly constant while the 
hydrometer is floating in it, and the temperature of the hydrometer must be the 
same as that of the water before it is immersed in it. In ordinary work on shore 
and in our latitudes this is the condition which it is most difficult to realise. In 
the Challenger it provided itself Nearly three out of the three and a half years 
that the voyage lasted were spent between latitudes 40° N. and 40° S. Here the 
temperature of the air is relatively high, but its diurnal variation is very slight. 
Moreover, the Challenger was a wooden ship, and the laboratory was lighted and 
ventilated by a large main-deck gun-port, the result of which was that, especially 
in the tropics, the temperature of the air was almost constant, day and night. 

The temperature of the surface water was usually slightly higher than that 
of the air, but only by a fraction of a degree, so that its specific gravity could be 
determined immediately after collection. Samples of water brought up from the 
bottom and the inferior depths arrived on board having a temperature much lower 
than that of the air, and it was impossible, even if it had been convenient, to 
proceed at once to the determinations of their specific gravity. A case containing 
eight large stoppered bottles was kept in the laboratory for the purpose of receiving 
these samples as they arrived, and they were kept in the laboratory until the next 
day, and their specific gravities were then determined one after the other. The 



24 



MR J. Y. BUCHANAN ON THE 



twenty-four hours' sojourn in the laboratory equalised their temperature and brought 
it to agree sensibly with that of the air of the laboratory and that of the hydrometer, 
which was always kept in the laboratory. By working according to this system, 
the specific gravities of the waters obtained from different depths at the same station 
were determined at the same time and at the same temperature, and their relative 
specific gravities at a common temperature were thus given directly by experiment. 
This is an important advantage, and it is often overlooked. 

A subjective precaution, but one of great importance for assuring accuracy of 
observation, was adopted at the beginning of the voyage and was never departed 
from. Before beginning to make hydrometric observations on the samples of water, 
or to carry out any other operation, such as the boiling out of the gases or the deter- 
mination of the carbonic acid, / locked the door of the laboratory, and it ivas not 
unlocked until the operation was finished. Consequently none of my colleagues, or 
anyone else in the ship, ever witnessed the determination of the specific gravity of 
the water, or any other of the operations carried on in the laboratory, at any time 
from the beginning to the end of the voyage. I found that exactness of observation 
was promoted by freedom from disturbance. 

Table VIII.* 

Giving Duplicate Observations of the same Sample of Water with the same 
Hydrom,eter differently weighted. 





Density observed with 






Density ob 


served with 




No. of 
Saiiijile. 






DilTereiice. 
OOIV-OOV. 


No. of 
Sample. 






Difference. 
OOIV-OOV. 












00 IV. 


oov. 






OOIV. 


OOV. 




120 


1-02412 


1-02411 


+ 1 


274 


1-02416 


1-02412 


-1-4 


127 


1-02414 


1-02409 


+ 5 


826 


1-02411 


1-02411 





135 


1-02406 


1-02413 


-7 


829 


1-02411 


1-02408 


+ 3 


139 


1-02407 


102414 


-7 


830 


1-02400 


1-02405 


-5 


181 


1-02428 


1-02427 


+ 1 


831 


1-02421 


1-02418 


+ 3 



After the Challenger Expedition I used in all my deep-sea work hydrometers with 
sets of weights designed for making multiple observations in each water. The principal 
stepping-stone between the Challenger hydrometer and that used in the investigations 
of this memoir was one in which the observations were made in triplicate, the difference 
between successive added weights being 0'25 or 0*3 gram. One reading was made near 
the middle of the stem, and the other two were made near the middle of the lower and 
the upper halves of the stem respectively. 

A very complete and important set of observations on this scheme was made in 
1885, on a voyage from Southampton to Buenos Ayres, and then from Valparaiso 
following the west coast of South America to Panama, and thence along the west coast 

* "Report on the Specific Gravity of Ocean Water," Physics and Chemistry, vol. i., Part II., Table VIII. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 25 

of North America to San Francisco. This pattern of hydrometer was also used a good 
deal in cable ships. 

In all work at sea three observations, or perhaps four, are quite sufficient. When 
four observations are made, their arithmetical mean has theoretically two-thirds of the 
value of the arithmetical mean of a series of nine. 

§ 8. As much misunderstanding seemed to exist not only regarding the qualifications 
required for the successful practice of the hydrometric method, but also, to some extent, 
of the principles on which the legitimacy of the method depends, I took occasion at the 
meeting of the Sixth International Geographical Congress, held in London in 1895, at 
which I read a paper entitled " A Retrospect of Oceanography during the last Twenty 
Years," to deal with some of these misapprehensions. The following passage may be 
quoted here with advantage : — 

" Many writers, in passing judgment on the hydrometer as an instrument for the 
determination of the density of liquids, have only in their minds the hydrometer whose 
indications are determined by comparison with another or standard instrument ; or by 
immersion in solutions the densities of which have been otherwise ascertained. These 
instruments have no greater value than that of more or less carefully constructed copies 
of a standard, the method and the principle of the construction of which is not always 
given. Rightly, therefore, they prefer the density as determined by weighing a vessel 
filled with the liquid and comparing it with the weight of distilled water of the same 
temperature filling the same vessel. The hydrometer which I constructed for the 
Challenger Expedition, and used during the whole of it, is not a hydrometer in the 
above sense : it does not give comparative results ; it gives absolute ones. By its 
means, the weights of equal volumes of the solution and of the distilled water of the 
same temperature are determined directly. It is neither more nor less than a pykno- 
meter, where the volume of liquid excluded up to a certain mark is weighed, instead ol 
that included up to a similar mark. In the pyknometer, the internal surface per unit 
of length of the stem can be made smaller than the external surface per unit of length 
of stem of the hydrometer. On the other hand, the volume of the hydrometer can 
safely be made many times larger than that of the pyknometer, the dimensions of which 
must always be kept small on account of the difiiculty of ascertaining the true tempera- 
ture of its contents, which must be guessed, because it cannot be measured directly. 
The temperature of another mass of liquid is measured, and the two are assumed to be 
identical. With the hydrometer, the liquid being in large quantity and outside of the 
instrument, its temperature can be immediately ascertained with every required accuracy. 

" Again, for every determination with the ordinary pyknometer, the weight of the 
liquid contained in it has to be determined by a separate operation of weighing. With 
the hydrometer, the weight of the liquid displaced, being always equal to its own, is 
determined once for all by repeated series of weighings, where every refinement is used 
to secure the true weight of the instrument. This weight can be increased at will by 
placing suitable small weights on the upper extremity of the stem. Their weight is 
TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 4 



26 



MR J. Y. BUCHANAN ON THE 



also most carefully determined once for all, so that at any moment the total weight of 
the displacing instrument is accurately known." * 



A- 



-D 




Section II, — The Principle and Construction op the Closed Hydrometer. 

§ 9. It will be convenient to follow in detail the preparation of the hydrometer 
for use. 

The instrument being closed, its true weight is constant. 

Let it be assumed that our experiments are actually made in vacuo, 
at the sea-level in lat. 45°. In these conditions the standard gram exerts 
a vertical pressure of 1 gram (true). 

We weigh the hydrometer and find its weight to be W grams. We 
now float it in distilled water contained in a suitable cylinder. In the 
construction of the hydrometer the internal load has been so adjusted 
that, when immersed in distilled water of the standard temperature T, 
which is to remain unaltered during the whole of the experiments, the 
surface of the water shall cut the stem in some line C, near its junction 
with the body of the instrument. Then the weight of the water displaced 
by the hydrometer is exactly W grams. 

Let pressure be now applied to the top of the stem, A, until it is 
completely immersed. Let the measure of this pressure be w grams. 
Then the weight of water displaced by the instrument when totally 
immersed at temperature T is (W + w) grams. 
\ / We assume that the stem is a uniform cylinder of circular section and 

X^ ^ terminated by a plane surface. If we apply pressure so as to immerse 
y^ ^v. the stem to the line D, which is midway between A and C, the pressure 

i^HI^BI required will be — grams ; and, if the portion of stem so immersed, CD, 

stands in any other ratio to the total length CA, the pressure required to 

produce the immersion will stand in the same ratio to w. 

Let the experiments be made in air of temperature T, and of pressure and humidity 

such that 1 cubic centimetre of it weighs 1 '2 milligram. When the experiment was 

made in the vacuum and the surface of the water cut the stem in C, the weight of the 

water so displaced was exactly equal to that of the hydrometer, namely, W grams. 

After air has been admitted, the surface of the water no longer cuts the stem 
exactly in C, but at a point a little lower. This difference between the lines of flotation 
is due to the fact that, while experimenting in vacuo, the portion of the stem which is 
not immersed in the water is immersed in a medium the density of which is insensible, 
whereas, after the air has been admitted, it is immersed in a medium of which 1 cubic 
centimetre weighs 1"2 milligram, and this exerts an upward pressure, in opposition to 



Report of the Sixth International Geographical Congress, held in London, 1895, p. 412. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 27 

gravity, at the rate of 1 "2 milligram per cubic centimetre of stem so immersed in air. 
This upward pressure lifts the hydrometer until it displaces a weight of water less than 
it did in vacuo by the weight of air which the exposed stem displaces after air has 
been admitted. 

Therefore, in ordinary laboratory practice, when the hydrometer floats in the liquid 
at any line C on the stem, the true weight of the liquid so displaced is equal to the true 
weight in vacuo of the hydrometer less the weight of the air displaced by the exposed 
portion of the stem. 

If s be the weight of the air displaced by the exposed portion of the stem, and W, 
as before, be the weight in vacuo of the instrument, the effective vertical pressure 
exercised by the hydrometer when floating in equilibrium on the water is 

H = (W-s), 

and this is the measure of its displacement in distilled water of temperature T under 
existing atmospheric conditions. 

§ 10. In instruments of the pattern, fig. 3, § 80, which I construct for use in dilute saline 
solutions I aim at a displacement of 180 grams distilled water. The stem is made from 
tubing selected with the greatest care so as to secure uniformity of calibre. Its total length 
is about 130 millimetres, and its external diameter is such that a length of 10 centimetres 
displaces something less than a cubic centimetre. This condition is satisfied if the 
glass-blower selects a suitable piece of tube having an external diameter of 3 to 3 '5 
millimetres by the callipers. If the diameter of the tube is exactly 3'56825 millimetres 
and its section circular, 10 centimetres of it will displace at 4° C. 1 cubic centimetre. 
The graduated portion of the stem occupies a length of 10 centimetres, which is divided 
into millimetres numbered at every centimetre from below upwards: 0, 1, 2, . . . 10. 
The zero is about 1 centimetre above the junction of the stem with the body, and the 
highest division, numbered 10, is found at a distance of about 2 centimetres below the 
top of the stem. The total length of the instrument should not exceed 33 centimetres. 

If the hydrometer floats in distilled water of temperature T so that the surface of 
the water cuts the stem at 5 millimetres above the zero of the scale (I express this 
shortly by saying, the hydrometer floats at 5), and the weight of air so displaced by 
the exposed stem is s^, then the true weight of water so displaced is 

The other conditions remaining the same, let the distilled water in the cylinder be 
replaced by a saline solution at temperature T. Let the hydrometer be floated in it ; 
the surface of the liquid will cut the stem or the body of the instrument at a lower 
level than the 5th millimetre on the scale. In order to immerse the instrument exactly 
to the 5th millimetre, we have to place a certain weight on the top of the stem. Let 
its weight in vacuo be w^ grams. Then the weight of liquid displaced by the system is 

H'j = W - s^ + W5 - dw^, 

where dw^ is the weight of the air displaced by the small added weight iv^. 



28 MR J. Y. BUCHANAN ON THE 

We have then two independent observations, namely, those of the weights of the 
distilled water and of the saline solution respectively, which occupy the same volume 
under identical conditions. The ratio of these two weights is the specific gravity 
of the heavier liquid referred to that of the lighter at the same temperature as 

unity. It is : — 

_ H's W - s^ + to^ - dio^ 
''' H,- W-s, 

Let us now repeat the double experiment, all the conditions remaining the same, 
except that, when the hydrometer has been immersed in the distilled water and floats 
at 5, a small weight v-^^ is added which immerses the hydrometer until it floats exactly 
at 15. Let % be the weight of air displaced by the exposed stem above the 15th 
division, and let dv^^ be the weight of air displaced by the small weight v-^^. Then the 
weight in vacuo of the distilled water displaced by the hydrometer below line 15 is 

Let the hydrometer be now immersed in the heavier liquid, and let weights be 
placed on the top of the stem until it floats exactly at 15. As before, the weight of 
this liquid so displaced is 

and the specific gravity of the liquid must be 

^1 1; — T^ — -:;^^ ; . 



■'IS' 



^15 W-Sis + U^s-C^Vis 



Now H5 and H'5 are the weights of equal volumes of distilled water and of a 
heavier liquid respectively, and H15 and H'15 are also weights of equal volumes of 
distilled water and of the same heavier liquid respectively : therefore in the two ratios 

TT/ TT/ 

=j^ and ==^ we have two independent values of the specific gravity of the heavier liquid 
under identical conditions, namely, 

S, = ll^ and S,, = |l^ 

As the specific gravity of each liquid has remained the same, these two independent 
determinations ought to give identical values for S : that is, S5 = S15. 

It is evident that we can increase at will the number of independent determinations 
of the specific gravity of the heavier liquid as referred to that of distilled water under 
constant conditions, and obtain from them a mean value of continually increasing 
exactness. 

It will be observed that the values of the specific gravity so obtained depend on 
our own observations alone. We have therefore the means of appraising their value 
exactly. Moreover, their value depends almost exclusively on determinations of weight : 
and this is the physical constant of a body which can be directly determined with 
perhaps greater precision than any other. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 29 

§ 11. Preparation of Accessory Weights. — We have now to consider the prepara- 
tion or manufacture of the small weights to be placed on the top of the stem in 
order to produce small increments of the immersion of the hydrometer. They are 
made of wire. This is wound into spiral cones for the heavier and into I'ings for 
the lighter weights. The lighter weights are made of aluminium and the heavier 
ones of brass. 

Generally a set of weights consists of aluminium spirals weighing 0"2, 0"5, and 1*0 
gram, and rings of the same metal weighing 0"2 and 0"1 gram, also rings 0"05 and 0"02 
gram. The brass weights are rings of 0"5 and 1*0 gram and spirals of 1, 3, 5, and 
7 grams. At every operation I aim at making a series of nine independent observ^a- 
tions of the displacement. In the first observation the lightest added weight is 
used and the reading (Kj) is near the zero of the scale. The succeeding observations 
are made while the added weight is increased by O'l gram between each observation 
of the series. The observations thus obtained are spread over the whole of the scale 
on the stem. 

The weights may be made so that their nominal weight is their true weight in vacuo, 
but, as they are always used in air, it is preferable to adjust them by balancing them 
against standard weights in air. The standard weights exert their nominal vertical 
pressure only in vacuo, at the sea-level in latitude 45° ; but we have assumed that we 
are in fact working at the sea-level in latitude 45°, therefore the nominal pressure of the 
standard weight is affected only by the density of the medium in which it is immersed. 
When we are actually working in a vacuum the density of the medium is insensible ; 
when we are working in air its density is ascertained by observation. Our standard 
weights, which have been verified at Kew, are made of brass (gilt) for weights of 
1 gram and upwards, and of platinum for weights under 1 gram. The weights destined 
for use on the stem of the hydrometer are also made of brass for those of 1 gram and 
upwards, and for those of 1 gram and under they are made of aluminium. There 
are gram weights and half-gram weights of both brass and aluminium. 

We will consider [a) the preparation of a gram weight of brass as balanced against 
a standard gram of brass ; and (6) the preparation of a gram weight of aluminium as 
balanced against a standard gram of platinum. 

(a) As we are dealing with only one kind of material, it is sufficient to equilibrate 
our weight of brass wire against the brass standard gram in air of known density to 
obtain a weight which in vacuo exerts a vertical pressure equal to that of the standard 
gram, and it must exert the same vertical pressure as does the standard gram in air of 
the same density. Taking the specific gravity of brass wire at 8*38, 1 gram of it displaces 
O'l 19 cubic centimetre of air, which, at 1*2 milligram per cubic centimetre, weighs 0*1428 
milligram. Therefore, when reckoning the effective pressure exerted by the brass weights 
placed on the top of the hydrometer in air of the density above specified, we make a 
deduction from their nominal weight in vacuo in the proportion of •1428 milligram per 
gram used. 



30 MR J. Y. BUCHANAN ON THE 

(6) Let us now consider the preparation of a weight of 1 gram in aluminium for the 
hydrometer, against a standard gram weight in platinum. We take the specific gravity 
of aluminium at 2*5 and that of platinum at 21. 

The volume of a gram of platinum is therefore 1/21 cubic centimetre, and it displaces 
this volume of air, which weighs 0"057 milligram. Therefore the standard platinum 
gram weighs in air 0*999943 gram or 0*057 milligram less than in vacuo. If w'e are 
working actually in the vacuum and we equilibrate the platinum gram with a mass of 
aluminium, both masses exert the same vertical pressure. But when we admit the air 
the platinum gram loses only 057 milligram of apparent weight, whereas the aluminium 

gram loses — ;:=0"48 milligram of weight, and its vertical pressure in air is only 
z '0 

0*99952 gram. 

A more useful result is obtained by equilibrating the platinum and aluminium 
in air. 

J^et the standard gram of platinum be placed on the one pan of the balance, and let 
a mass of aluminium which in vacuo weighs 1 standard gram be placed on the other 
pan. The two masses which, in vacuo, would exactly balance each other, now appear 
to have different weights. By immersion in the air the platinum gram has lost 0*057 
milligram and the aluminium gram has lost 0*48 milligram. Let aluminium be 
added to the aluminium weight until the balance shows equilibrium. The amount so 
added weighs in air 0*423 milligram. The vertical pressures exerted in the air by the 
masses of platinum and aluminium respectively are then equal. But this pressure is 
still short of the standard pressure of 1 gram by 0*057 milligram. Let this weight 
of aluminium be added to the mass of aluminium already on the pan. When this 
addition has been made, the total mass of aluminium will exert in air, weighing 1 '2 
milligram per cubic centimetre, a vertical pressure of 1 gram true. No account has been 
taken of the buoyancy of the last two additions to the mass of aluminium, because its 
effect is insensible on our balance. 

In practice the aluminium weights used in any experiment never exceed 1 gram 
by more than one or two tenths ; therefore, if they have been simply balanced 
against the corresponding platinum weights in air, the deduction for buoyancy 
is insensible ; and we have seen that, if the brass weights have been prepared 
against brass standards in air, the deduction for buoyancy is at the rate of 0*14 milli- 
gram per gram when 1 cubic centimetre of air weighs 1*2 milligram per cubic 
centimetre. 

§ 12. Exposed Stem. — Let us consider the effect on the resulting value of the specific 
gravity of a liquid when the correction for the buoyancy of the exposed stem is 
applied or is neglected. The effect of buoyancy will evidently be the greater, the 
greater the length of the exposed stem. Let us take the case of the hydrometer 
suitably loaded, floating at mm., or the lowest division on the stem, both in the 
distilled water and m the solution. The volume of the exposed stem is 1*25 cubic 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 31 

centimetres in both cases, and the air which it displaces, at 1'2 milligram per cubic 
centimetre, weighs I "5 milligram. 

In order to avoid complication, we suppose that the necessary "added weights" 
have been added to the internal load of the closed hydrometer, and that the dis- 
placing weight of the hydrometer quoted for each immersion is its true weight in 
vacuo, and that nothing which can affect the immersion of the instrument in the 
liquid is immersed in air excepting the exposed portion of the stem itself. This 
disengages the effect produced by the buoyancy of the stem from that of every 
other cause. 

Let the weight of the hydrometer so floating at mm. in distilled water be 180 "25 
grams ; and let its weight when floating also at mm. in the solution be 185 "25 grams; 
then, neglecting the buoyancy of the stem, the specific gravity of the solution is 
1 '027739. But the effect of buoyancy is to reduce the effective weight in both cases by 
1*5 milligram, so that the specific gravity of the solution corrected for buoyancy of 

stem is =1 "027740. Therefore, when the whole of the stem is exposed, its 

180-2485 ^ 

buoyancy affects the resulting specific gravity to the extent of only a unit in the sixth 

decimal place. 

§ 13. Determination of the Weight of the Hydrometer. — For this purpose the 
hydrometer is placed on the right-hand pan of the balance in an upright position, and 
is brought to equilibrium with weights and rider on the left-hand pan. The hydrometer 
is then removed and equilibrium again established by means of standard weights. 
These are then replaced by the hydrometer and equilibrium re-established by shifting 
the rider of the counterpoise if it has been disturbed. The hydrometer is again 
removed and replaced by standard weights until equilibrium is established. In this 
way four independent weighings by replacement by standard weights are obtained. 
The temperature of the air is noted, also the temperature of the wet-bulb thermo- 
meter and the height of the barometer. Three such series of weighings are made 
on different days when the meteorological conditions are different. Each series is 
treated by itself. In order to obtain the vacuum correction we require to know the 
weight of the air displaced by the hydrometer and by the weights respectively. 
The difference of these two weights, the net buoyancy, is the correction to be added 
to the apparent weight of the hydrometer. 

We take as an example of the method the determination of the weight in vacuo 
of hydrometer No. 17. 

1st Determination. 
5th March 1894. 

Barometer = 740'86 mm. 

Temperature, dry bulb = 6*15°, wet bulb= 5*1° C. 

Whence the vapour tension is 6*03 mm., and the weight of 1 litre of this air is 

1-2288 gram. 



32 MR. J. Y. BUCHANAN ON THE 

Four weighings of the hydrometer by replacement with standard weights were 
made in air. The weights found were : — 

180-7141 grams. 
180-7137 „ 
180-7136 „ 
180-7137 „ 



Mean = 180-7138 „ 

As the hydrometer floats without added weights in distilled water with only a 
part of the stem exposed, we take its weight in air, 1807138 grams, as expressing, to 
first approximation, its volume in cubic centimetres. The correction for net buoyancy, 
that is, the difference between the weight of air displaced by the hydrometer and 
that displaced by the weights, is (taking dry air at G'lS" C. and 741 mm.) 0"1961 gram, 
whence the first approximation to the weight in vacuo of the instrument is 
1807138 + 0-1961 = 180-9099 grams. 

It was found that by the addition of 1*698 gram to the weight of the hydrometer 
it floated totally immersed in distilled water at 6*15° C. ; that is, if the weight were 
diminished ever so little the top of the stem became exposed, and if it were increased 
ever so little the instrument began to sink to the bottom. 

Taking now the first approximation to the weight in vacuo, 180"9099 grams, and 
adding r6980 gram, we have the sura 182*6079 grams. This is the first approximation 
to the weight in vacuo of the mass of distilled water which is displaced by the whole 
hydrometer at a temperature of 6*15° C. Taking the volume of 1 kilogram of water 
at 6"15° C. to be 1000*034 cubic centimetres, we find the volume of the hydrometer 
at 6*15° C. to be 182*6139 c.c. ; and this is the volume of air which it displaces at 
6*15° C. We have found that 1 litre of the air in the balance-room at the time 
weighed 1*2288 gram. Therefoi'e the exact weight of the air displaced by the 
hydrometer when being weighed was 182*6139 x 00012288 = 0*22439 gram, and 
taking the weights as consisting of brass of the density 8*38, we find the weight of 
air displaced by them to be 0*02650 gram, whence the net buoyancy = 0* 19789 
gram, and the true weight in vacuo of the hydrometer is 180*7138 + 0*1979 
- 180*9117 grams. 

The weight in vacuo was determined on two other days, namely 24th April and 
2nd June 1894; on each of these days four determinations were made of the weight 
in air, which on the first of these days weighed 1*2081 gram per litre, and on the 
second 1*2037 gram per litre. The weights in vacuo deduced from these observations 
were 180*9109 and 180*9113 grams respectively. The mean of the three determina- 
tions is 180*9113 grams, which is accepted as the final value of the weight in vacuo of 
hydrometer No. 17. The weights of the other hydrometers, Nos. 21 and 3, were 
determined in the same way ; the particulars are collected in the following table : — 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 33 



Deter- 
mination 
No. 



Date. 



Mean Weight 

of Hydro- 
meter in Air. 

Grams. 



Number 

of 
Weigh- 
ings in 
Air. 



Barometer. 



Thermometer. 



Dry 
Bulb. 



Wet 
Bulb. 



Vapour 
Pressure. 

Milli- 
metres. 



Weight 

of 1 litre 

of this 

Air. 

Grams. 



Exact net 
Buoyancy. 

Grams. 



Exact Weight 
in vacuo of 
Hydrometer. 

Grams. 



Hydrometer No. 17. 



1 

2 
3 


5/3/1894 

24/4/1894 

2/6/1894 


180-7138 
180-7165 
180-7175 


4 
4 
4 


740-86 
740-46 
747-01 


6° 15 
10-50 
1390 


5"l0 

8-90 

11-50 


6-03 

7-70 
8-89 


1-2288 
1-2081 
1-2037 


0-1979 
1944 
0-1938 

Mean = 


180-9117 
180-9109 
1809113 




180-9113 



Hydrometer No. 21. 



24/11/1893 i 187-5771 

5/12/1893 I 187-5764 

27/ 4/1894 I 187-5809 



4 
4 
4 


74905 
7r)5-40 
74365 


8-25 

9-85 

13-40 


6-70 

800 

1105 


6-56 
7-07 
8-65 


1-2327 
1-2361 
1-2006 


0-2060 
0206ti 
0-2006 

Mean = 



187-7831 
187-7830 
187-7815 



187-7825 



Hydrometer No, 3. 



9/3/1894 

10/5/1894 

2/6/1894 



178-1785 


4 


731-36 


9°-45 


7-70 


6-98 


1-1982 


0-1904 


178-1786 


4 


740-20 


14-10 


10-75 


7-97 


1-1924 


0-1895 


178-1776 


4 


746-89 


13-95 


11-50 


8-87 


1-2032 


0-1912 
Mean = 



178-3689 
178-3681 
178-3688 



178-3686 



No. 17. 
Final weight in vacuo accepted for each hydrometer, 180'91] 3 grams. 



No. 21. No. 3. 

187-7825 grams. 178-3686 grams. 



§ 14. As the displacement in distilled water figures in all the determinations of 
density, we begin by making a number of series of observations of the displacement 
of the hydrometer in it at the standard temperature chosen. 

We take as an example the case of hydrometer No. 17 in distilled water of 1 5 "00° C. 
as it is given in Table A^. 

Table A-y. — This table gives in detail the data from which is derived the total 
weight displaced by hydrometer No. 17 when floating at the 50-mm. mark in distilled 
water at 15*00° C. Each line in the table is distinguished by a letter — a, b, c, etc. 
In line a we have the number and particulars of the hydrometer used. In line b 
is given a reference to the laboratory note-book in which the original observations 
were entered, followed by the date of the experiment, line c. Lines d and i give 

TRANS. ROY. SCO. EDIN., VOL. XLIX., PART L.(NO. 1). 5 



34 MR J. Y. BUCHANAN ON THE 

the times at which the hydrometer is immersed and removed, while the temperatures 
of the water at these times are given in lines e and j, with their mean, T, in line k ; 
the range of temperature over which the series of observations was carried out is 
shown in line /. Although it is intended that the temperature shall be uniform 
during an experiment, it sometimes happens that it varies, and this has to be 
provided for in the table. Line f^ gives the headings iv, the weights used to sink 
the hydrometer in the liquid, and R, the reading on the scale of the instrument 
corresponding to these weights. 

Lines f-^ to f^ give, under w, the values of the weights, and under R, the 
corresponding scale readings, obtained during the series of observations. The value 
of the mean added weight, w, is given in line g, while the mean reading, R, is 
shown in line h. 

The departure of the mean reading from 50 mm., 50 — R, is entered in line m ; it is 
given in the headings as dr. In line n we have the weight which is equivalent to df, 
expressed as dw^ (see § 15). 

The weight required to immerse the hydrometer to the 50-mm. mark, w-\-dWr, 
irrespective of temperature corrections, is shown in line o, being that weight which 
would cause the instrument to float with the scale division at 50 mm. in the plane of 
the surface of the liquid, at the mean temperature, T. The difi"erence of the mean 
temperature, T, from the standard temperature, T, is given in the line p, and is 
expressed as T — T = c?i ; the weight corresponding to the difference dt is entered in 
line q; it is expressed as dwt (see § 16). 

The total corrected added weight required to immerse the hydrometer to the 
50-mm. mark at the standard temperature, T, is w + dwr-'r dwt, and is given in 
line r. 

The total weight of liquid displaced by the hydrometer when floating in the 
liquid at the 50-mm. mark at the standard temperature, T, is entered in line s, and 
is equal to the weight of the instrument in vacuo plus w + dwr + diVf 

Having explained the meaning of the lines, we will proceed to inspect the results 
of the observations in Table A^. 

After preliminary trial, the first weight added to the hydrometer at the commence- 
ment of a series of observations is chosen so that the mean of the nine series of 
immersions or scale readings produced by successive added weights, each increasing 
by O'l gram, shall approximate closely to 50 mm. It is evident that the initial, or 
first, added weight might be different for each series of observations. In the ten series 
of observations detailed in Table Aj, however, the first added weight in each case 
was 0'525 gram, and therefore the nine added weights are given only once, under w, 
in lines /^ to /g of the first column. Each of the ten succeeding columns contains a 
complete series of observations of the immersions produced by the nine added weights, 
and the steps in the calculation of the total weight of the hydrometer when floating 
at the 50-mm. mark at the standard temperature. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 35 

§ 15. Correction for Departure of the Mean Reading from 50 mm. — If the mean 
reading, R, be exactly 50 mm., then the weight which must be added to the 
hydrometer to immerse it to the 50-mm. mark is the mean added weight, w. 

If, however, the mean reading, R, be less or greater than 50, the mean added 
weight must be increased or diminished by the weight, dw^, which would increase or 
diminish the immersion by the difference df between the observed mean reading, R, and 
50. The calculation of this correction, dw^, is best explained by taking the series of 
observations, XVII. 73, Table A^, as an example. In this case the mean reading, R, is 
5072, and (if = 50 - 5072 = -072 (Knew). 

The immersion in the whole series of observations is increased by 89 '2 mm. 
of the stem by an addition of O'SOO gram. Hence an increase of 1 mm. in the 

stem immersion is caused by the addition of - = 0*00897 gram, and a difference of 

^ 89-2 * 

072 mm. in the immersion must be produced by the weight 0"00897 x — 072 — — 0'0064 
gram, which is the required correction, dw^ (line n). 

As the mean reading, R, is in this case greater than 50, the weight required to 
immerse the hydrometer to 50 mm. must be less than the mean added weight, w, by 
the amount of the correction, dw^, and is, therefore, 0'925 — 0'0064 = 0"9186 gram 
(line o). If the mean reading, R, were less than 50 mm., the correction, dwr, would 
be calculated in the same manner, but would require to be added to the mean added 
weight. 

§ 16. Correction for Temperature. — The weight required to be added to the hydro- 
meter to cause it to float at the 50-mm. mark in distilled water of the mean observed 
temperature of 15'01° C, as found above, is 0*9186 gram. A correction {dwt, line q) 
must now be applied to reduce the displacement observed at the mean temperature, 
T, to the standard temperature, T, which is in this case 15*00° C. Before this can 
be done we must determine the value of dw^ for 0*01° C. at 15*00° C. This is 
found as follows. A series of observations is made with the hydrometer in distilled 
water at various temperatures, the results of which are expressed in a curve, having 
displacements as abscissae and temperatures as ordinates. Suppose we wish to 
find the temperature correction at, say, 23*00° C, we proceed as follows. Draw 
horizontal lines through T = 23"5° and T = 22*5° C, cutting the curve at a and h 
respectively. From a drop a perpendicular on ch, meeting it at c. Then the length 
ac represents 1° C, while ch is the difference in the total displacement for this 1° 
difference in the temperature. Knowing the value of the abscissa OX in grams per 
unit length, say grams per millimetre, we measure accurately the length ch and 
multiply it by this constant. This gives us the value of dwt for 1° difference of 
temperature at 23° C. The value of dwt per 0*01° C. is simply the former figure 
divided by 100. This process is repeated at each of the temperatures at which 
observations are being made. 

The value of dwt in grams per 0*01° C. at 15*00° C. has been taken as 0*00026. 



36 



MR J. Y. BUCHANAN ON THE 



The temperature at the commencement of the observations in our example was 1 5*00°, 
and at the end 1 5 "02°, the mean being 15*01°. 

The departure of the mean temperature from 15*00° is, T — T = (]?i = 1501 — 15"00 
= 0"01° (line p). Therefore the amount by which the added weight must be increased 
for the difference dt is dwt = 0*00026 x 1 = 0*00026 gram (line q). The mean tempera- 
ture observed during the time the observations were being made was higher than the 
standard, so that in this case we must add the correction for temperature to the 
added weight required to immerse the stem to 50 mm. at 15*00° C. 



y 

2600 



2300". . 



1950' 



I 

i 



15 00°- ■ 





181'hOOO 



J • 000475 gm per 0/ "C 



00035 gm per 01 "C 




00050 gm per ore 



Total Displac/ng Weight 

H 1 



■00026 gm per ore I 



5000 -6000 7000 

Temperature Curve for HyorometerN^//. 



■+- 



80 OQ 



Finally, by adding together the mean added weight, w, the weight, dw^, for 
the difference of the mean reading, R, from 50 mm., and the weight, dWi, for 
the difference of the mean observed temperature, T, from the standard temperature, 
T, we obtain the weight ii; + c?w;^ + c^'M;j= 0*925 + (- 0*0064) + 0*00026 = 0*91886 gram 
given in line r, which must be added to the hydrometer to immerse it to the 
50-mm. mark in distilled water of 15*00° C, and by adding this weight to that 
of the hydrometer in vacuo we obtain 180*9113 + 0*91886 = 181*83016 grams 
(line s), which is the total weight of water displaced by the hydrometer under 
these conditions. 

^17. The data for the determination of the total weights of hydrometer No. 17 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 37 

when floating at the 50-mm. mark in distilled water of 19 "50°, 23 "00°, and 26*00° C. 
are given in Tables Ag, Ag, and A^ respectively. 

The mean of the values entered in line s of each of the Tables Aj, Ag, A3, and A4 is 
accepted as the correct total weight of hydrometer No. 17 when floating at the 50-mm. 
mark in distilled water at the temperatures 15'0°, 19'5°, 230°, and 26'0'' C. respectively ; 
they are collected in Table B with the corresponding values for hydrometers 
Nos. 21 and 3. 



[Tables. 



38 



MR J. Y. BUCHANAN ON THE 



Table 

Hydrometer 

Details of Determination of the Total Weight of the Hydrometer 



Particulars of h)'dron)eter, . 
Reference to laboratory note-book, 

Date of expeiimeiit, 

Time at start, .... 

Temperature at start, . 

Added weight, w, and reading, R, 

First added weight or reading. 

Second ,, ,. 

Third 

Fourth ,, ,, 

Fifth 

Sixth ,, 

Seventh ,, ,, 

Eighth 

Ninth ,, ,, 

Mean added weight, w. 

Mean reading, R, 

Time at finish, 

Final temperature. 

Mean temjterature, T, 

Range of temperature, _ 

Difference of mean reading from 50 mm. (50 - 'K^df), 

Weight equivalent to displacement, df { = dWr), . 

Weight required to immerse hydrometer to 50-mm. mark at mean tempera 

ture, T, {w + dwr), _ 

Difference of mean temperature, T, from standard temperature (T-T — di) 
Correction for difference, rfi ( = rfw<), ....... 

Weight required to immerse hydrometer to 50-mm. mark at standard tem 

perature, T, (w + dwr + dwt), 
Total weight displaced by hydrometer when floating at 50-mm. mark at 

standard temperature, T, 



a 


Hydi 


. b 


... 


c 




. d 


... 


e 




■ fo 


IV 


■ f^ 


0-525 


■ U 


0-625 


■ U 


0-725 


■ U 


0-825 


. u 


0-925 


• h 


1-025 


• /r 


1-125 


. u 


1-225 


• U 


1-3-25 


• g 


0-925 


. h 




X 




■ 3 




. k 




. I 




m 




n 









• V 




• ? 




r 




t s 





ometer No. 17. 
II. 161 

1903 

Feb. 28 

10.58 a.m. 

15-00° 

R 

5-8 
17-0 
27-8 
39-1 
50-2 
61-3 
72-5 
84-8 
95-0 

50-39 

11.18 a.m. 

15-00° 

15-00° 

0-00 

-0-39 

-0-C035 

0-9215 



0-9215 
181-8328 



J. Y. B., 1893 
II. 165 
1903 
Feb. 28 
11.30 a.m. 
15-02° 
R 
6-0 
17-1 
28-2 
39-3 
50-3 
61-3 
72-0 
84-0 
95-2 

50-37 
11.45 a.m. 
14-98° 
15-00* 
0-02° 
-0-37 
-0-0033 
0-9217 



0-9217 
181-8330 



Table 

Hydrometer 

Details of Determination of the Total Weight of the Hydrometer 



Particulars of hydrometer, 
Reference to laboratory note-book, 

Date of experiment, . 

Time at start, . 

Temperature at start, 

Added weight, w, and reading. 

First added weight or reading. 

Second ,, ,, 

Third 

Fourth ,, ,, 

Fifth ,, ,, 

Sixth ,, ,, 

Seventh ,, ,, 

Eighth 

Ninth ,, _,, 

Mean added wei^lit, w, 

Mean reading, R, 

Tiine at finish, . 

Final temperature, . 

Mean temperature, T, 

Range of temperature. 

Difference of mean reading from 50 mm. (50 -K = df), 

Weight equivalent to disidacement, df ( =dwr), 

Weight required to immerse hydrometer to 50-mm. mark 
at mean temperature, T, {w + dwr), 

Difference of mean temperature, T, from standard tempera- 
ture (T-T=rfO. 

Correction for difference, d?(=dw/), 

Weiglit required to immerse hydrometer to 50-mm. mark 
at standard temperature, T, (w + dwr + diot), 

Total weight disi)laced by hydrometer when floating at 
50-mm. mark at standard temperature, T, 



a 




b 


... 


c 




d 


... 


e 


•■. 


/o 


w 


t\ 


0-4 


U 


0-5 


A 


0-6 


h 


0-7 


h 


0-8 


U 


0-9 


U 


1-0 


h 


1-1 


,/» 


1-2 


9 


0-8 


h 




I 




3 




k 




I 


. .* 


m 




n 







... 


P 




? 




r 




s 





IV. 137 


IV. 143 


IV. 149 


1903 


1903 


1903 


March 23 


March 23 


March 23 


11.35 a.m. 


2.0 p.m. 


3.0 p.m. 


19-58° 


19-58° 


19-50° 


R 


R 


R 


5-1 


6-0 


5-5 


16-2 


17-2 


17-0 


27-5 


28-0 


28-0 


38-5 


39-0 


39-0 


50-0 


500 


49-5 


61-0 


60-8 


61 


72-0 


72-5 


72-3 


83-3 


83-0 


83-5 


94-5 


94-5 


94-8 


49 79 


.50-11 


50-06 


11.52 a.m. 


2.15 p.m. 


3.15 p.m. 


19-40° 


19-48° 


19-50° 


19-49° 


19-53° 


19-50° 


0-18° 


0-10° 


0-IjO 


0-21 


-0-11 


-0-06 


0-0018 


-0-0009 


-0-0005 


0-8018 


0-7991 


0-7995 


-0-01° 


0-03° 


... 


-0-00026 


0-00078 




0-8015 


0-7998 


0-7995 


181-7128 


181-7111 


181-7108 



Hydrometer No. 17. J.Y.B., 1893 

J IV. 149 IV. 155 IV. 161 

1903 1903 

March 24 March 24 

12 p.m. 1.51 p.m. 

19-53° 19-50° 

R R 

5-3 5-5 

16-5 17-0 

27-8 28-0 

39-0 39-2 

50-2 50-1 

61-9 610 

72-1 72-5 

83-8 83-8 

95-0 95-0 

50-17 50-23 

12.15 p.m. 2.5 p.m. 

19-51° 19-50° 

19-52° 19-50° 

0-02° 0-00 

-0-17 -0-23 

-00015 -0-0020 

0-7985 0-7980 

0-02° 

0-0005 

0-7990 0-7980 

181-7103 181-7093 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 39 



No. 17. 

when floating at the 50-mm. mark in Distilled Water at 15 '00° C. 



Weight in vacuo 


= 180-9118 grams. T=15-00°C 














IV. 29 


IV. 33 


IV. 37 


XVII. 73 


XVII. 75 


XVII. 77 


XVII. 79 


XVII, 81 




1903 


1903 


1903 


1904 


1904 


1904 


1904 


1904 




March G 


March 6 


March 6 


Feb. 2 


Feb. 2 


Feb. 2 


Feb. 2 


Feb. 2 




11.0 a.m. 


11.58 a.m. 


1.32 p.m. 


11.15 a.m. 


11.35 a.m. 


12.23 p.m. 


2.20 p.m. 


2.40 p.m. 




15-01° 


15-00° 


15-00° 


15-00° 


15-02° 


15-00° 


15-00° 


15-00° 




R 


R 


R 


R 


R 


R 


R 


R 




6-1 


6-1 


6-1 


6-0 


6-3 


6-3 


6-3 


6-3 




17-2 


17-1 


16-5 


17-2 


17-5 


17-3 


17-5 


17-5 




27-5 


27-5 


28-2 


28-5 


29-0 


28-8 


28-5 


29-0 




39-8 


39-0 


39-5 


39-8 


40-0 


39-9 


39-9 


40-0 




50-5 


50-2 


50-5 


50-8 


51-0 


51-0 


51-0 


51-1 




62-3 


62-0 


62-0 


62-0 


62-0 


62-0 


62 


62-1 




72-5 


72-2 


73 


73-0 


73-1 


73-0 


73-0 


73-1 




84-0 


84-0 


84-5 


84-0 


84-2 


84-1 


84-1 


84-1 




96-0 


95-3 


96-0 


95-2 


95-5 


95-3 


95-3 


95-3 




50-65 


50-37 


50-70 


50-72 


50-95 


50-85 


50-84 


50-94 




11.14 a.m. 


12.10 p.m. 


1.45 p.m. 


11.33 a.m. 


11.50 a.m. 


12.35 p.m. 


2.35 p.m. 


2.52 p.m. 




14-99° 


15 00° 


15-00° 


15-0-2° 


15-05° 


15-00* 


15-00° 


15-02° 




1500° 


15 00° 


15-00° 


15-01° 


15-035° 


1500° 


15-00° 


15-01° 




0-02° 


0-00 


0-00 


0-02° 


0-03° 


0-00 


0-00 


0-02° 




-0-65 


-0-37 


-0-70 


-0-72 


-0-95 


-0-85 


-0-84 


-0-94 




-0-0058 


-0-0033 


-0-0062 


-0-0064 


-0-0085 


- 0-0076 


-0-0075 


-00084 




0-9192 


0-9217 


0-9188 


0-9186 

0-01° 
0-0002 


0-9165 

0-035° 
0-0009 


0-9174 


0-9175 


0-9166 

o-oi" 

0-0002 




0-9192 


o'-'9217 


0-9188 


0-9188 


0-9174 


0-9174 


0-9175 


0-9168 




181-8305 


181-8330 


181-8301 


181-8301 


181-8287 


181-8287 


181-8288 


181-8281 



A,. 

No. 17. 

when floating at the 50-mm. mark in Distilled Water at 19 '50° C. 



Weiyht in mcMO = 180-9113 grams. 


T=19-50° 


C. 
















IV. 167 


VI. 7 


VI. 13 


VI. 19 


VI. 25 


VI. 27 


VI. 29 


VI. 31 


VI. 33 


VI. 35 


VI. 37 




1903 


1903 


190J 


1903 


1903 


1903 


1903 


1903 


1903 


1903 


1903 




Murch 24 


March 25 


March 25 


March 25 


March 26 


March 26 


March 26 


March 26 


Match 26 


March 26 


March 26 




2.40 p.m. 


12.12 p.m. 


2.55 p.m. 


3.47 p.m. 


12.0 p.m. 


12 18 p.m. 


12.35 p.m. 


1.0 p.m. 


2.48 p.m. 


3.3 p.m. 


3.30 p.m. 




19-50° 


19-53° 


19-50° 


19-50° 


19-50° 


19-50° 


19-50° 


19-50° 


19-50° 


19 50° 


19-51° 




R 


R 


R 


R 


R 


R 


R 


R 


R 


R 


R 




5-8 


5-3 


5-8 


5-8 


5-3 


6-0 


5-5 


5-8 


5-5 


5-8 


5-5 




17-0 


16-5 


17-0 


170 


16-5 


170 


16-5 


17-0 


16-5 


17-0 


16-5 




28-0 


27-8 


28-0 


27-8 


27-5 


27-5 


27-8 


28-0 


27-5 


28-1 


27-9 




39-0 


39-0 


39-2 


39-1 


39-0 


39-0 


39-0 


39-1 


39-0 


39-3 


39-0 




50-0 


50-1 


50-3 


50-2 


50-0 


50-0 


50-0 


50-2 


49-8 


50-2 


50-3 




60-8 


61-3 


61-0 


61-3 


61-0 


61-0 


61-0 


61-3 


61-0 


61-2 


61-5 




72-0 


72-2 


72-5 


72-5 


72-5 


71-5 


72-0 


72-5 


72-3 


72-5 


72-5 




83-5 


84-5 


84-0 


83-8 


83-5 


83-5 


83-2 


83-5 


83-5 


83-8 


83-5 




95-0 


95-0 


95-0 


95-0 


94-8 


95-0 


94-5 


95-0 


94-5 


95-0 


94-8 




50-12 


50-19 


50-31 


50-27 


50-01 


50-05 


49 94 


50-26 


49-95 


50-32 


50-16 




2.52 p.m. 


12.25 p.m. 


3.10 |..m. 


4.2 p.m. 


12.15 p.m. 


12.33 p.m. 


12.50 p.m. 


1.15 p.m. 


3.2 p.m. 


3.20 p.m. 


3.45 p.m. 




19-50° 


19-55° 


19-53° 


19-53° 


19-50° 


19-50° 


19 51° 


19-66° 


19-50° 


19-55° 


19-53° 




19-50° 


19-54° 


19-515° 


19-51.5° 


19-50° 


19-50° 


19-50° 


19-53° 


19-50° 


19-525° 


19-52° 




0-00 


0-02° 


0-03° 


0-03° 


0-00 


0-00 


0-01° 


0-06° 


0-00 


0-05° 


0-0-2° 




-0-12 


-0-19 


-0-31 


-0-27 


-0-01 


-0-05 


0-06 


-0-26 


0-05 


-0-32 


-0-16 




-0-0010 


-0-0017 


-0-0027 


-0-0024 


0-0000 


-0-0004 


0-0005 


-0-0023 


0-0004 


-0-0U28 


-0-0014 




0-7990 


0-7983 


0-7973 


0-7976 


0-8000 


0-7996 


0-8005 


0-7977 


0-8004 


0-7972 


0-7986 






0-04° 


015° 


0-015° 




... 




0-03° 


... 


0-025° 


0-02° 




... 


0-0010 


0004 


0-0004 








0-0008 




0-0008 


0-0005 




0-7990 


0-7993 


0-7977 


0-7979 


0-8000 


0-7996 


0-8005 


0-7985 


0-8004 


0-7979 


0-7991 




181-7103 


181-7106 


181-7090 


181-7092 


181-7113 


181-7109 


181-7118 


181-7098 


181-7117 


181-7092 


181-7104 



40 



MR J. Y. BUCHANAN ON THE 



Table 

Hydrometer No. 17. 

Details of Determination of the Total Weight of the Hydrometer 



Particulars of hydrometer, 


a 






Hydrometer No. 17. 


Reference to laboratory note-book, 




• 












h 




XXVm. 59 
1905 


XXV 11 1. 61 

1905 


XXVIII. 63 
1905 




Uatfi of experiment, . 
















c 




June 6 


June 6 


June 6 




Time at start, . 


















d 




10.52 a.m. 


11.42 a.m. 


12.25 p.m. 




Temperature at start, 


















c 




23-05° 


23 00° 


23-03° 




Added weight, w, nnd reading, 


R, 
















/o 


w 


R 


R 


R 




First added weight or reading. 


















/l 


0-25 


2 2 


1-8 


2-2 




Second ,, ,, 


















/a 


0-35 


13-5 


12-2 


12-5 




Third 


















/s 


0-45 


24-8 


23-8 


24-5 




Fourth 


















/4 


0-55 


35-8 


35-0 


35-8 




Filth 


















/s 


0-65 


47 


46-2 


46-8 




Sixth 


















/e 


075 


68-0 


57-2 


57-2 




Seventh ,, ,, 


















A 


0-85 


690 


68-5 


68-8 




Eighth 


















/a 


0-95 


80-0 


79-5 


79-2 




Ninth ,, „ 


















/d 


1-05 


90-8 


90-5 


91-2 




Mean added wej^ht, w, 


















9 


0-65 










Mean reading, R, 


















h 


... 


46 '-jg 


46-08 


46-47 




Time at finish, . 


















i 


... 


11.15 a.m. 


12.00 p.m. 


12.40 p.m. 




Final temperature, _. 


















3 




22-85° 


22-90° 


23-02° 




Mean temperature, T, 


















k 




22-95° 


22-95° 


23 025° 




Range of temperature. 


















I 




0-20° 


0-10° 


0-01° 




Difference of mean reading from 50 mm. (50-R = <^r), 










m 




3-21 


3-92 


3 -,^3 




Weight ei|Uivalent to displacement, df {=dwr), 










n 


... 


0-0290 


0-0354 


0-0319 




Weiglit required to immerse liydrometer to ."iO-mm. mark at mean tempera- 







0-6790 


0-6854 


6819 




ture, T, {w + dwr). 














Difference of mean temperature, T, from standard temperature {T-T^dl), 


P 




-0-05° 


-0-05° 


0025° 




Coriection for diti'ereace c?i (=rft«(), 


q 




-0-0015 


-0-0015 


00007 




Weight required to immerse hydrometer to 50-mm. mark at standard tem- 


r 


... 


0-6775 


6839 


0-6826 




perature, T, { = w + dwr + diLH), 














Total weight displaced by hydrometer when iloating at 50-mm. mark at 


s 




181-5888 


181-5952 


181-5938 




standard temperature, T, 















Table 

Hydrometer No. 17. 

Details of Determination of the Total Weight of the Hydrometer 



Particulars of hydrometer. 
Reference to laboratory note-book. 

Date of experiment. 

Time at start. 

Temperature at start, 

Added weight, w, and reading, R 

First added weight or reading. 

Second ,, ,, 

Third 

Fourth ,, ,, 

Fifth 

Sixth ,, 

Seventh ,, ,, 

Eighth ,, ,, 

Ninth ,, ,, 

Mean added weight, w, . 

Mean readinjr, R, . 

Time at finish, 

Final tcmpeiature, 

Mean temperature, T, 

Kange of tempeiature, . 

Dill'erence of mean reading from 50 mm. (50- \i = df) 

WeiL'ht equivalent to di.splacenicnt df ( =dw,), . _ 

Weight required to iramor.se hydrometer to 50-mm. mark at mean temperature, T, {iv + dWr). 

Dill'eieiice of me in temperature, T, from standard temperature (T - T = rf?) 

Correction for dilletence d/I (=rfw(), ....... 

Weisiht required to immerse hydrometer to .'iO-mm. maik at standard temperature, T, 

( =w + dwr + dwt), 
Total weight displaced by hydrometer when floating at 50-ram. mark at standard 

tem[)erature, T, 



a 
b 

c 
d 
e 

/o 
/i 

A 

fz 

/4 

h 
h 

fn 
fs 

h 

9 
h 
i 

J 

k 
I 

m 
n 



P 
9. 

r 



Hydrometei- No. 
VIIL 47 

1903 

July 2 

3.20 p.m. 

26-00° 



17. 



w 
0-175 
0-275 
0-375 
0-475 
0-575 
0-675 
0-775 
0-875 
0-975 
0-575 



R 
6-0 

17-1 
28-5 
39-8 
50-8 
62-0 
73-2 
84-3 
95-6 

50-81 
3.35 p.m. 

26-10° 

26-05° 
0-10° 

-0-81 

-0-0072 
0-5678 
0-05° 
0-0018 
0-5696 

181-4809 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 41 



J. Y. B., 1893. 

when floating at the dO-mm. mark in Distilled Water at 2 3 "00° C. 



J. Y. B., 189 


3. Weight 


in rflsmo = 180-9113 gram 


5. T = '23-00 


°C. 












XXVIII. 65 


XXVIII. 67 


XXVIII. 69 




XXX. 107 


XXX. Ill 


XXX. 141 


XXX. 143 


XXX 145 


XXX. 147 




1905 


1905 


1905 




1905 


1905 


1905 


1905 


1905 


1905 




June 6 


June 6 


June 6 




July 12 


July 12 


July 17 


July 17 


July 17 


July 17 




1.40 p.m. 


2.17 pm. 


4.3 p.m. 




10.35 a.m. 


11.20 a.m. 


1.54 p.m. 


2.22 p.m. 


3.15 p.m. 


4.0 p.m. 




23-00° 


23-00° 


23 00° 




22-80° 


23-10° 


23-00° 


23-00° 


23-00° 


23-00° 




R 


R 


R 


w 


R 


R 


R 


R 


R 


R 




2-0 


2-2 


1-5 


0-275 


3-2 


5-2 


6-2 


6-2 


5-2 


. 6-2 




12-2 


12-8 


11-5 


0-375 


15-2 


16-2 


17-2 


17-2 


16-2 


17-2 




23-8 


23-5 


23-2 


0-475 


26-2 


27-2 


27-8 


28-0 


27-5 


28-0 




35-2 


35 


34-2 


0-575 


37-0 


37-8 


38-2 


38-2 


38-2 


38-2 




45-0 


46 


45-2 


0-675 


48-2 


47-8 


49-0 


49-2 


48-5 


49-2 




55-8 


57-5 


55-8 


0-775 


59-2 


60-2 


600 


60-2 


60-0 


60-2 




68-2 


68-5 


67-8 


0-875 


70-2 


71-2 


71-2 


71-2 


71-2 


71-2 




79-2 


78-8 


78-5 


0-975 


81-2 


82-2 


82-2 


82-2 


82-2 


82-5 




90-2 


90-0 


89-5 


1-075 
0-675 


93-2 


94-2 


94-2 


94-2 


93-2 


94-2 




45-73 


46-03 


45-25 




48-177 


49-111 


49 555 


49-622 


49-133 


49-655 




2.0 p.m. 


2.30 p.m. 


4.23 p.m. 




10.57 a.m. 


11.40 a.m. 


2.8 p.m. 


2.40 p.m. 


3.33 p.m. 


4.10 p.m. 




23 00° 


23-00° 


'23-00° 




23 00° 


23-22° 


2300' 


23-00° 


23-00° 


23-00° 




23-00° 


23-00° 


23-00° 




22-90° 


23-16° 


-23-00" 


23 00° 


23-00° 


23-00° 




00 


0-00 


0-00 




0-20° 


0-12° 


0-00 


0-00 


0-00 


0-00 




4-27 


3-97 


4-75 




1-823 


0-889 


0-445 


0-378 


0-867 


0-345 




0-0386 


0-0360 


0-0430 




0-0164 


0-0080 


0-0040 


0-0034 


0-0078 


0-0031 




0-6886 


6860 


0-6930 




0-6914 

-0-10° 
-0-0031 


0-6830 

0-16° 
0-0049 


0-6790 


0-6784 


0-6828 


0-6781 




0-6886 


0-6860 


0-6930 




0-6883 


0-6879 


0-6790 


6-6784 


6-6828 


6-6781 




181-5999 


181-5973 


181-6043 




181-5996 


181-5992 


181-5903 


181-5897 


181-5941 


181-5894 



A4. 

J. Y. B., 1893. 

ivhen floating at the 50-mm. mark in Distilled Water at 26*00° C. 



J. Y. B., 1893. 


Weight wwcMO = 180-9113 grams. T = 26-00' 


'C. 










VIII. 49 


VIII. 51 


VIII. 53 


VIII. 55 


VIII. 57 


VIII. 59 


VIII. 61 


VIII. 63 




1903 


1903 


1903 


1903 


1903 


1903 


1903 


1903 




July 2 


July 2 


July 2 


July 3 


July 3 


July 3 


July 3 


July 3 




3.35 p.m. 


3.55 p.m. 


4.15 p.m. 


12.15 p.m. 


12. 30 "p.m. 


12.45 p.m. 


1.5 p.m. 


11.50 a.m. 




26-10° 


26-10° 


•26-00° 


26-01° 


26-00° 


26-00° 


26-00° 


26-00° 




R 


R 


R 


R 


R 


R 


R 


R 




6-8 


6-8 


7-0 


6-3 


6-3 


6-3 


6-3 


6-3 




17-8 


17-8 


18-2 


17-0 


17-0 


17-1 


17-1 


16-3 




29-2 


29-3 


29-5 


27-8 


28-5 


28-5 


28-2 


27-0 




40-^ 


40-5 


40-5 


39-8 


39-2 


39-0 


39-3 


39-2 




51-5 


51-5 


51-8 


50-8 


50-8 


50-5 


50-0 


50-1 




62-8 


62-5 


63-0 


61-3 


61-8 


61-8 


61-8 


61-3 




74-0 


74-0 


74-0 


73-0 


73-0 


72-3 


72-2 


73-0 




85-0 


85-0 


85-0 


84-0 


84-2 


84-0 


84-5 


84-0 




96-2 


96-2 


96-2 


95-0 


95-3 


95-5 


95-5 


95-0 




51-53 


51-51 


51-68 


50-55 


50-67 


50-55 


50-54 


50-24 




3.50 p.m. 


4.10 p.m. 


4.30 p-m. 


12.28 p.m. 


12.45 p.m. 


1.0 p.m. 


1.15 p.m. 


12.10 p.m. 




26-10° 


26-00° 


26-00° 


25-95° 


25-98° 


26-00° 


26-00° 


25-80° 




26-10° 


26-05° 


26-00° 


25-98° 


25-99° 


26-00° 


26-00° 


25-90° 




0-00 


0-10° 


0-00 


0-06° 


0-02° 


0-00 


0-00 


0-20° 




-1-53 


-1-51 


-1-68 


-0-55 


-0-67 


-0-55 


-0-54 


-0-24 




-0-0137 


-0-0135 


-0-0152 


-0-0049 


-0-0060 


-0-0049 


-0-0048 


-0-2100 




0-5613 


0-5615 


0-5598 


0-5701 


0-5690 


0-5701 


0-5702 


5729 




0-10° 


0-05° 




-0-02° 


-0-01° 


• *. 




-0-10° 




0-0036 


0-0018 


..• 


-0-0007 


-0-0003 






-0 0036 




0-5649 


0-5633 


0-5598 


0-5693 


0-5687 


0-5701 


0-5702 


0-5693 




181-4762 


181-4746 


181-4711 


181-4806 


181-4800 


181-4814 


181-4815 


181-4806 



TRA.NS. ROY. SOG. EDIN., VOL. XLIX., PART I. (NO. 1). 



42 



MR J. Y. BUCHANAN ON THE 



Table 

HydroTneter 

Details of Determination of the Total Weight of the Hydrometer 



Particulars of h)'drometer, 
Reference to laboratory note-book, 

Date of experiment, . 

Time at start, . 

Temperature at start. 

Added weight, w, and reading, R, 

First added weight or reading, 

Second , , , , 

Third 

Fourth , , , 

Fifth 

Sixth 

Seventh , , , , 

Eighth 

Ninth ,, _ „ 

Mean added weight, w, 

Meau reading, R, 

Time at finish, . 

Final temperature, _. 

Mean temperature, T, 

Range of temperature, 

Difference of mean reading from 50 mm. (50 - R=t/r), 

Weight equivalent to displacement df ( =d'w, 

Weight required to immerse hydrometer to 50-mm. mark 
at mean temperature, T, {w + divr), 

Difference of mean temperature, T, from standard tempera- 
ture (T-T=d<), 

Correction for difference c?i ( = (;?W() 

Weight required to immerse hydrometer to 50-mm. mark 
at standard temperature, T, {=w + d^Vr + dwt), 

Total weight displaced by hydrometer when floating at 
50-mm. mark at standard temperature, T. 



a 






Hydrometer No. 21. J. 


Y. B., 1893. 


b 




II. 17 


IV. 139 


IV. 145 


IV. 151 


IV. 157 






1903 


1903 


1903 


1903 


1903 


c 




Feb. 10 


March 23 


March 23 


March 23 


March 24 


d 


... 


1.15 p.m. 


12 p.m. 


2.20 p.m. 


3.15 p.m. 


12.25 p.m. 


e 




19-8° 


19-55° 


19-55° 


19-50° 


19 51° 


h 


w 


R 


R 


R 


R 


R 


A 


0-4 


4-2 


40 


3-5 


3-8 


3-5 


h 


0-5 


160 


15-3 


15-1 


15-5 


15-1 


h 


0-6 


28-0 


27-1 


27-0 


27-2 


26-2 


h 


07 


39-1 


39-0 


38-5 


39-0 


38-5 


h 


0-8 


500 


610 


50-5 


50-8 


50-5 


h 


0-9 


62-5 


62-3 


62-2 


62-5 


62-0 


A 


1-0 


74-2 


74-2 


74-0 


74-3 


73-5 


U 


1-1 


86-0 


86-0 


86-0 


86-0 


85-2 


U 


1-2 


97-3 


97-5 


97-5 


97-5 


97-5 


g 


0-8 












h 




50-81 


50-71 


50-47 


50-73 


50-22 


i 




1.28 p.m. 


12.15 p.m. 


2.35 p.m. 


3.30 p.m. 


12.40 p.m. 


J 


... 


19-50° 


19-49° 


19-45* 


19-51° 


19-50° 


k 


... 


19-65° 


19-52° 


19-50* 


19-505° 


19-505° 


I 


... 


0-30° 


0-06° 


0-10° 


0-01° 


0-01° 


in 


... 


-0-81 


-0-71 


-0-47 


-073 


-0 22 


n 




-0-0069 


-0-0060 


-0-0040 


-0-0062 


-0-0018 







0-7931 


0-7940 


0-7960 


0-7938 


0-7982 


P 




o-is" 


0-02° 




0-005° 


0-005° 


q 




0-0040 


0-0005 




0-0001 


0-0001 


r 




0-7971 


0-7945 


07960 


07939 


0-7983 


s 




188-5796 


188-5770 


188-5785 


188-5764 


188-5808 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 



43 





^5- 

No. 21. 






















when Jloating at the 50-mm. mai 


^k in Distilled Water at 19-50° C. 






Weight in i>«ffio = 187 '7825 grams. 


T = 19-50° 


c. 
















IV. 163 


IV. 169 


VL 9 


VI. 15 


VI. 21 


VI. 39 


VI. 41 


VI. 43 


VI. 45 


VI. 47 


VI. 49 




1903 


1903 


1903 


1903 


1903 


1903 


1903 


1903 


1903 


1903 


1903 




March 24 


Marcli 24 


March 25 


March 25 


March 25 


March 27 


March 27 


March 27 


March 27 


March 27 


March 27 




2.8 p.m. 


2.53 p.m. 


2.40 p.m. 


3.15 p.m. 


4.3 p.m. 


2.0 p.m. 


2.15 p.m. 


•2.40 p.m. 


3.0 p.m. 


3.18 p.m. 


3.38 p.m. 




19-50° 


19-50° 


19-50° 


19-50° 


19-53° 


19 -50° 


19-50° 


19-53° 


19-50° 


19-50° 


19-50° 




R 


R 


R 


R 


R 


R 


R 


R 


R 


R 


R 




3-8 


3-8 


3-1 


4-0 


4-0 


3-5 


3-3 


4-2 


4-0 


4-0 


4-0 




15-5 


15-5 


15-0 


15-5 


16-0 


15-3 


15-2 


16-0 


15-8 


15-8 


15-5 




27-2 


27-5 


26-8 


27-5 


28-0 


26-8 


27-2 


28-0 


27-5 


27-5 


27-5 




39-0 


39-1 


38-5 


39-1 


39-3 


38-5 


39-0 


39-3 


391 


39-0 


39-2 




50-8 


51-0 


50-5 


51-0 


61-0 


50-5 


51-0 


51-0 


51-0 


51-0 


51-0 




62-5 


62-8 


61-8 


62-5 


62-8 


62-3 


62-5 


63-0 


62-8 


62-2 


63-0 




74-3 


74-5 


74-0 


74-2 


74-8 


74-2 


74-5 


75-0 


74-5 


74-5 


74-5 




86-0 


86-0 


860 


86-0 


86-3 


86-0 


86-0 


86-5 


85-8 


86-0 


86-2 




97-5 


98-0 


97-5 


97-5 


98-0 


97-5 


98-0 


98-0 


97-5 


98-0 


98-0 




50-73 


50-91 


50-35 


50-81 


51-13 


.50-51 


50-74 


51-22 


50-88 


50-88 


50-98 




2.23 p.m. 


3.10 p.m 


2.55 p.m. 


3.30 p.m. 


4.18 p.m. 


2.13 p.m. 


2.30 p.m. 


2.55 p.m. 


3.15 p.m. 


3.35 p.m. 


3.55 p.m. 




19-50° 


19-5,5° 


19-50" 


19-50° 


19-59° 


19-50° 


19-55° 


19-51° 


19-50° 


19-52° 


19-58° 




19-50° 


19-525° 


19-50° 


19-50° 


19-56° 


19-50° 


19-525° 


19-52° 


19-50° 


19-51° 


19-54° 




0-00 


0-05° 


0-00 


0-00 


0-06° 


0-00 


0-05° 


0-0-2' 


0-00 


0-02° 


0-08° 




-0-73 


-0-91 


-0-3.5 


-0-81 


-1-13 


-0-51 


-0-74 


-1-22 


-0-88 


-0-88 


-0-98 




-0-0062 


-0-0077 


-0-0030 


-0-0069 


-0-0096 


-0-0043 


-0-0063 


-0-0104 


-0-0075 


-0-0075 


-0-0083 




0-7938 


07923 


7970 


0-7931 


07904 


0-7957 


07937 


07896 


07925 


0-7925 


07917 






0-025° 




... 


0-06° 


... 


0-025° 


0-02° 




0-or 


0-04° 






0006 






0-0016 




0-0006 


0-0005 




0-0002 


0-0011 




07938 


0-7929 


0-7970 


0-7931 


0-7920 


0-7957 


0-7943 


7901 


0-7925 


0-7927 


0-7928 




188-5763 


188-5754 


188-5795 


188-5756 


188-5745 


188-5782 


188-5768 


188-5726 


188-5750 


188-5752 


188-5753 



44 



MR J. Y. BUCHANAN ON THE 



Table B. 
Observed Total Weights of Hi/drometers tvhen floating .at the bQ-mm. mark 
in Distilled Water at various Temperatures. 



Hydrometer. 


Temperature. 


Mean Total Weight. 
Grams. 


Number of 

Series of 

Observations. 


Maximum Departure 
from the Mean. 


Probable Error of 
Arithmetical Mean. 


17 


15-00 


181 8304 


10 


0049 


+ 

0-0006 




19-50 


181-7105 


16 


0-0023 


0-0002 




23-00 


181-5952 


12 


00093 


0-0010 




26-00 


181-4785 


9 


0-0074 


00008 


21 


15-00 


188-7012 


12 


0-0085 


0-0007 




19-50 


188-5767 


16 


0-0040 


0-0003 




23-00 


188-4570 


15 


0-0107 


0-0011 




26-00 


188-3363 


9 


0-0031 


0-0005 


3 


15-00 


179-3116 


12 


0-0117 


0-0012 




19-50 


179-1987 


12 


0-0040 


0-0005 




23 00 


179-0757 


4 


0-0006 


0-0001 




26-00 


178-9697 


16 


0-0038 


0-0002 



Section III. — Determination of the Specific Gravity of a Saline Solution. 

§ 18. When determining the specific gravity of a saline solution by the hydrometric 
method, it is necessary first to find the weight which must be added to the hydrometer 
to immerse it to the 50-mm. mark when floating in the saline solution at the 
chosen standard temperature. This added weight is found by a series of observations 
in exactly the same manner as with the hydrometer in distilled water (see § 14 ei seq.). 

The details of three series of observations with hydrometer No. 17 in a solution 
of \ gram-molecule of csesium chloride in 1000 grams of water at 19 "50^ are given 
as an example in Table C. This table is arranged in the same manner as Table A^. 

The correction for the difference of the mean immersion from 50 mm. is calculated 
in the same manner as has been explained in connection with the determination of the 
displacement of hydrometer No. 17 in distilled water. 

Taking the series XX. 79 as an example, we find that an increase of O'S gram in 
the added weight increases the stem immersion from 6*0 to 93*5 mm., that is, 
87 '5 mm. ; hence each mm. increase in the stem immersion is caused by an addition 
of 0*800/87'5 = 0'00914 gram. The difference of the mean reading from 50 mm. in 
the series in question is 50 — 49'62 = 0*38, which is equivalent to an added weight of 
0-00914 X 0-38 - 0-0034 gram. 

As the mean reading is less than 50 mm., the hydrometer is not sufficiently 
immersed in the solution ; therefore the added weight is too small and must be in- 
creased. The weight to be added is 0'0034 gram; the resultant weight which must 
be added to the hydrometer to immerse it to the 50-mm. mark in the solution at 
the mean observed temperature is therefore 3-7 -H 0'0034 = 3 7034 grams (line o). 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 45 



As in the three series detailed in Table C the mean observed temperature is 
identical with the standard temperature, 19 "50°, no temperature correction is required, 
and the weight 37034 grams is that which is, in the first case, required to immerse 
the hydrometer to the 50-mm. mark in the saline solution of 19-50°. 

Adding this weight to that of hydrometer No. 17 in vacuo, we get 180'9113 
+ 37034= 184'6147 grams, entered on line s ; this is the total weight of the hydro- 
meter when floating at the 50-mm. mark in the solution under experiment. 

Dividing this number by the displacement of the hydrometer when floating at the 
50-mm. mark in distilled water of the same temperature, 19'5° C, namely, 18r7105 
grams, line s', Table C, we obtain 1 '01 5982 as the specific gravity of the solution. 
The specific gravity calculated from each series of observations is given at the foot of 
the corresponding column, in line t. 

Table C. 

Hydrometer No. 17. 

Details of the Determination of the Total Weight of the Hpdrometer when floating 
at the 50-m?>i. mark in a solution of ^ gram-molecule of Ccesium Chloride in 
1000 grams of water at 19*50° C, and the Specific Gravity of the Solution. 



Particulars of hydrometer, . 


a 


Hyd. No. 17. 


J. Y. B., 1893 


Weight in vacuo 


= 180-9113 grms. 


Reference to laboratory note-books, 


b 




XX. 79 


XX. 87 


XX. 91 


Date of experiment, .... 


c 




April 22, 190i 


April 22, 1904 


April 22, 1904 


Time at start, ..... 


d 




2.00 p.m. 


3.18 p.m. 


3.50 p.m. 


Temperature at start. 


t 




19-50° 


19-50° 


19-50° 


Added weight, w, and reading, R, 


/o 


w 


R 


R 


R 


First added weight or reading, R, 


/l 


3-3 


6-0 


5-5 


5-5 


Second 


f1 


3-4 


16-5 


16-5 


16-5 


Third 




h 


3-5 


28-0 


27-5 


27-5 


Fourth 




A 


3-6 


38-5 


38-3 


38-5 


Fifth 




h 


3-7 


49-8 


49-5 


49-5 


Sixth 




U 


3-8 


60-5 


60-3 


60-3 


Seventh ,, ,, , 




fi 


3-9 


71-3 


71-1 


71-1 


Eighth „ 




fs 


4-0 


82-5 


82-0 


82-5 


Ninth 




/o 


41 


93-5 


93-0 


93-5 


Mean added weight, w. 


9 


3-7 








Mean reading, R, . . . . 


h 




49-62 


49-30 


49-43 


Time at finish, ..... 


i 




2.15 p.m. 


3.35 p.m. 


4.05 p.m. 


Final temperature, .... 


J 




19-50° 


19-50° 


19-50° 


Mean temperature, T, ... 


k 




19-50° 


19-50° 


19-50° 


Range of temperature. 


I 




0-00 


0-00 


0-00 


Difference of mean reading from 50 


m 




0-38 


0-70 


0-57 


mm. (50-R = dr), 












Weight equivalent to displacement dr 


n 




0-0034 


00063 


0-0052 


{ = dwr), 












Weight required to immerse hydrometer 







3-7034 


3-7063 


3-7052 


to 50-mm. mark at mean temperature, 












T, {w + dio,), 












Difi'erence of mean temperature, T, 


P 


... 


000 


0-00 


0-00 


from standard temperature, T, (T - T 












= dt), 












Correction for difference dt ( = dwt), 


9 




0-00 


0-00 


0-00 



46 



MR J. Y. BUCHANAN ON THE 



Table C — continued. 



Weight required to immerse hydro- 


r 




3-7034 


3-7063 


3-7052 


meter to 50-mm. mark at standard 












temjjerature, T, ( = w + dio,. + dw^, 












Total weight of solution displaced by 


s 


• • • 


184-6147 


184-6176 


184-6165 


hydrometer when immersed to 50-nim. 












mark at standard temperature, T, 












Corresponding weight in distilled water 


s 


... 


181-7105 


181-7105 


181-7105 


Specific gravity of solution-^ 


t 


... 


1-015982 


1-015998 


1-015992 



§ 19. Influence of the Meniscus. — It may be here pointed out that the weight 
which causes the immersion of the hydrometer is its own weight plus that of the liquid 
meniscus which it carries on the stem above the line of flotation. It is impossible to 
measure or weigh this exactly ; but we are concerned only with the question if, 
and to what extent, it can affect the exactness of the determination of the specific 
gravity of our solutions. 

The weight of the meniscus depends on the surface-tension of the liquid, and this 
has been determined for distilled water and for a few saline solutions. The results are 
given in Tables Nos. 124 to 129 of the Smithsonian collection of Physical Tables, of 
which a new edition has been recently published.* In these tables only one salt 
belonging to either of the enneads MR or MROg is included, namely, chloride of 
potassium. For the solutions of the remaining seventeen salts there are no experi- 
mental data. We are therefore unable to state exactly the effect which would be 
produced on the specific gravities of their solutions if the weights of the meniscuses of 
the distilled water and of the solution had been taken into account. Nevertheless, it 
is worth while to study the influence of the meniscus on the specific gravity of the 
solutions of the salts quoted in the table, because light will thereby be thrown on 
the probable extent of its influence on the specific gravity of the solutions of the 
salts of the two enneads. 

When we determine the specific gravity of a solution, we first immerse the hydro- 
meter in distilled water, and add (if necessary) small weights to the top of the stem 
until it floats at a certain division on the stem, say 50 ; and we call the total displacing 
weight H. We then immerse the hydrometer in the solution, having the same 
temperature as the distilled water had, and add weights to the top of the stem until 
the instrument floats in the solution at the same division, 50 ; and we call the total 

TT/ 

displacing weight H'. Then the specific gravity of the solution is S = -^ • Here we 

have not taken account either of the meniscus of the distilled water or of that of the 
solution. But each of these meniscuses exerts the pressure of its own weight and 
contributes to the displacing weight of the instrument when it floats at 50 in the 
distilled water and the solution respectively. Let h = the weight of the meniscus of 

* Smithsonian Physical Tables^ 5tli revised edition, 19 lU. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 47 
distilled water, and let h' = that of the solution, then the two displacing weights are 

XT/ I JJ 

H + h and H' + h\ and the specific gravity is S^ = ^ — j- ■ The data given in the 

Smithsonian Tables, Nos. 124 and 126, enable us to arrive at the value of h for distilled 
water. Table No. 124, and at those of h! for certain solutions of the salts included 
in Table No. 126. In the following table the effect of introducing the weight of 
the meniscus in the hydrometric determination of the specific gravity of some of the 
solutions contained in Smithsonian Table No. 126 has been calculated. The tabular 
specific gravity (col. h) furnished by the Smithsonian Tables is taken to represent the 
hydrometric specific gravity arrived at without taking into account the influence of the 
meniscus. From this, and using the surface-tension of distilled water, and that of the 
solutions furnished by the Smithsonian Tables given in columns d and e, the hydro- 
metric specific gravity of the same solution, having regard to the weight of the 
meniscus, is calculated and entered in column I. The details of this calculation are as 
follows. In column a we have the formula of the salt in solution, in column h the 
specific gravity of the solution. In column c is the temperature, and in column e is the 
surface-tension of the solution at that temperature. The surface-tension of distilled 
water at the same temperature is given in column d. In columns d and e the values 
of the surface-tension are expressed in dynes per centimetre. Accepting 1/981 gram as 
the pressure which balances the force of 1 dyne, the weight, in milligrams, of liquid 
lifted by 1 centimetre of glass is obtained by multiplying the entries in the fourth 
column by TO 19. If the circumference of the stem of our hydrometer were exactly 
1 centimetre, this would be the weight of the meniscus. 

The circumference of the stem of hydrometer No. 17, the one which has been most 
frequently used, is 1*062 centimetre; therefore, to get the weight in milligrams of the 
meniscus supported by it, we must multiply the entry in column d by 1*019x1 "062 
= 1-0822. 

Consider the first line of the table. The salt dissolved is NaCl. Taking the 
total weight of hydrometer No. 17 when floating at the 50th division in distilled water 
to be 1817105 grams (col. h), and multiplying this number by r036 (col, 6), the 
tabular specific gravity of the least concentrated solution of NaCl quoted in the table, 
we obtain 188 "2521 (col. i) as the displacing weight of this hydrometer when floating 
at the 50th division in NaCl solution of the specific gravity r0360. The surface- 
tension of distilled water at 20° C. is given (col. d) as 72-8 dynes per centimetre. 
Multiplying this by r0822, we obtain 78*8 milligrams (col. / ) as the weight of the 
meniscus of distilled water. Similarly, multiplying 7 7 '6, the surface-tension of the NaCl 
solution in dynes per centimetre, by r0822, we obtain 84'0 milligrams (col. g) as the 
weight of the meniscus of this solution supported by the stem of hydrometer No. 17. 
We have, then, after taking account of the influence of meniscus, for the displacing weight 
of the hydrometer when floating in distilled water, 1817105 -f- 0-0788 = 181-7893 grams 
(col. j), and for that of the hydrometer when floating at the same division in NaCl solu- 



48 



MR J. Y. BUCHANAN ON THE 



tion, 188-2521 +0-0840= 188-3361 grams (col. k). Dividing this number by 181*7893, 
we obtain for the specific gravity of the solution, amended for the weight of meniscus, 



^h- 



188-3361 
181-7893 



= 1-036013 (col. Z). 



The difference between the two specific gravities is 1-8 in the fifth place. This 
difference is higher than the average. In the case of the nearly saturated solution of 
NaCl, sp. gr. = ri932, the difference is only 1 in the fifth place. In the corresponding 
solutions of KCl the difference in the dilute solution is 1, and in the concentrated 
solution 2, in the fifth place. In the case of CaClg both solutions are of considerable 
concentration ; the difference in the more concentrated is 2, and in the less concentrated 
r7, in the fifth place. In the case of MgClg the difference in the dilute solution is 1, 
and in the concentrated solution 0-4, in the fifth place. It is therefore evident that 
the specific gravity of solutions determined by the hydrometric method is not affected 
by an error due to the influence of the meniscus which calls for correction. 

The specific gravity of the less concentrated NaCl solution is 1'036, that of average 
oceanic sea-water is not greater than r027 ; therefore the influence of the meniscus in 
sea-water of highest concentration would be not greater than 1 in the fifth place. 
Tn all my hydrometric ivork on sea-water the infiuence of the meniscus has been dis- 
regarded, because I recognised from the beginning that the ratio of the iveights of the 
meniscus must be the same as that of the masses of the liquids displaced by the same 
im,mersed portion of the hydrometer, unless their volumes differed considerably, and of 
this there was no evidence. 



a. 


h. 


e. 


d. 


il-0822c^. 


l-0822e. 

g- 


h. 


hxb. 

i. 


h+f. 
J- 


i+g. 


k/j. 
I. 


l-b. 
m. 


Salt 
dis- 
solved. 


Without 
Meniscus. 


Temp. 
°C. 

20 
20 
15 
15 
15 
15 
19 
19 


Surface- , height in 
tension, M,-n;„^o„,c «f 


Without Meniscus. 


With Meniscus. 


With 
Meniscus. 


Difference. 


Specific 
Giavityof 
Solution. 


dynes 
per cm. 


Meniscus. 


Weight of 
Hydrometer in 


Weight of 
Hydrometer in 


Specific 
Gravity of 
Solution. 


H,0. 


Solu- 
tion. 


HaO. 


Solu- 
tion. 


HgO. 


Solution. 


H2O. 


Solution, 


NaCl 

NaCl 

KCl 

KCl 

MgCl^ 

MgCl^ 

CaCl2 

CaCl2 


1-03600 
1-19320 
1-04630 
1-16990 
1-03620 
1-23380 
1-27730 
1-35110 


72-8 
72-8 
73-5 
73-5 
73-5 
73-5 
72-9 
72-9 


77-6 
85-8 

78-2 
82-8 
78-0 
90-1 
902 
95-0 


78-8 
78-8 
79-5 
79-5 
79-5 
79-5 
78-9 
78-9 


84-0 
92-8 
84-6 
89-6 
84-4 
97-5 
97-6 
102-8 


181-7105 
181-7105 
181-7105 
181-7105 
181-7105 
181-7105 
181-7105 
181-7105 


188-2521 
216-8168 
190-1237 
212-5831 

188-2884 
2241943 
232-0987 
245-5090 


181-7893 

181-7893 
181-7900 
181-7900 
181-7900 
181-7900 
181-7894 
181-7894 


188-3361 
216-9096 
190-2083 
212-6726 
188-3728 
224-2918 
232-1963 
245-6118 


1-036013 
1-193190 
1-046310 
1-169880 
1036210 
1-233796 
1-277283 
1-351080 


+ 0-000013 
-0-000010 
-0-000010 
- 0-000020 
+ 0-000010 
-0-000004 
-0-000017 
-0-000020 



§ 20. Serial Determination of the Specific Gravity of a Saline Solution. — The 
method by which the specific gravity of a saline solution is determined with 
the hydrometer is given in another place (see § 18), and it is there shown how the 
total displacement of the instrument when floating in the solution up to the 50-mm. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 49 

mark, and at the standard temperature, is obtained. In that method the mean values 
of the added weights and the corresponding scale readings are utilised to obtain 
one value of the specific gravity of the solution. The conditions of the method are 
that exactly the same volume of the hydrometer shall be immersed in the solution 
and the distilled water, the temperature of both being the same, and the total weight 
required to cause the instrument to float at 50 mm. in the saline solution divided 
by the total weight required to float it at the 50-mm. mark in distilled water gives 
the specific gravity of the solution. 

In an experiment the hydrometer is loaded with nine successive weights in- 
creasing by O'l gram in each step, and the reading of the stem is observed after each 
addition. It is evident that if we take the first of the readings and find the weight 
that would have to be added to the hydrometer when it was floating at the same 
level in distilled water, the ratio of the two total weights would give a value for 
the specific gravity of the solution. In a similar way we might use the second 
scale reading as the basis for another computation of the specific gravity ; and, in 
general, each of the nine values of the scale readings, obtained by the addition of 
successive increments of O'l gram to the stem of the hydrometer, furnishes a separate 
basis for determining the specific gravity. Proceeding in this manner, the accompany- 
ing table has been compiled. 



Table giving Specific Gravity o/" 1/32 i?6C/+ 1000 grms. H^O at 19 '5° C. 

(a) Time : 2.00 p.m. 



tv. 


t. 


dwi. 


W. 


R. 


Wh,o. 


S. 


0-91 


19-5000 




181-821300 


5^9 


181-314344 


1-002796 


1-01 


-4975 


-000065 


•921235 


170 


•413622 


798 


1-11 


•4950 


130 


182-021170 


28-2 


•514932 


789 


1-21 


•4925 


195 


•121105 


39-3 


•613497 


795 


1-31 


•4900 


260 


•221040 


50-5 


•715553 


781 


1-41 


•4875 


325 


•320975 


61-5 


-814891 


783 


1-51 


•4850 


390 


•420910 


72-3 


-911848 


798 


1-61 


•4825 


455 


•520845 


83^5 


182-010671 


800 


1-71 


•4800 


520 


■620780 


94-9 


•112-200 
Mean 


792 
r002793 



(b) Time : 2.31 p.m. 



IV. 


t. 


dWf. 


W. 


R. 


W„,o. 


S. 


0-91 


19-5000 




181-821300 


6^5 


181-319701 


1-002766 


roi 


•5062 


-000161 


•921461 


179 


•421691 


54 


111 


•5124 


322 


182^021622 


29-0 


-522019 


52 


1^21 


•5186 


484 


•121783 


40^0 


•619795 


64 


1-31 


•5248 


645 


-221944 


51-1 


•721003 


61 


1-41 


-5310 


806 


•322105 


62-1 


•8-20278 


60 


1-51 


•5372 


967 


•422266 


73-1 


-918880 


67 


1-61 


-5434 


-001128 


•522427 


84-2 


182-016885 


77 


1-71 


•5500 


1300 


•622588 


95-5 


-117518 
Mean 


73 
1-002764 



TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 



50 



MR J. Y. BUCHANAN ON THE 









(c) Time: 3.02 p.m. 






w. 


t. 


dw,. 


W. 


R. 


W„,o. 


S. 


0^91 

roi 

Ml 
121 
1-31 

r4i 

1^61 
1-71 


19-5000 


0000000 


181-821300 
•921300 

182-021300 
-121300 
-221300 
-321300 
-421300 
•521300 
•621300 


6-2 
17-5 
29^0 
400 
510 
62^0 
73^0 
84^0 
95^2 


18r317012 
•418105 
•522019 
•619795 
•720107 
•819382 
•917984 

182^015092 
•114828 
Mean 


1-002782 
73 
50 
61 
57 
59 
66 
81 
81 

1-002768 









{d) Time : 3.30 p.m. 






w. 


t. 


diOi. 


W. 


R. 


Wh,o. 


S. 


0-91 


19-5200 


-000520 


181-821820 


6^5 


181-319701 


r002769 


101 






-921820 


17^8 


-420794 


61 


111 






182-021820 


29^0 


-522019 


53 


121 






-121820 


40^0 


-619795 


64 


1-31 






•221820 


bl-0 


-720107 


59 


1-41 






•321820 


62^0 


•819382 


63 


1-51 






•421820 


73^0 


•917984 


66 


1-61 






•521820 


84^2 


182^016885 


74 


1-71 






•621820 


95^3 


•115724 
Mean 


99 
1^002765 



Dealing first with sub-table (a), we see that the temperature at the commencement 
was 19'5° C, and at the end of the series of observations 19"48°, giving a difference of 
0"02° fall during the time that the observations were being made. As nine separate 
observations were made, we can assume that the temperature fell by equal amounts 
between each two observations, so that if we take as the first temperature 19*5°, and 
then successively subtract 0'00"25° for each reading, the last will be taken at 19*48° C. 
As the temperature falls, the density of the solution rises, so that in order to com- 
pensate for the fall of temperature, the weight, diVi, corresponding to this departure of 
temperature from the standard must be subtracted from the total observed weight of 
the hydrometer. The values of dwi are placed in the column headed diVt. Under iv 
are given the small weights which were placed on the stem of the hydrometer in order 
to immerse the instrument. Under W is given the total weight of the hydrometer, 
corrected for temperature, that causes it to float at the scale reading, R, at 19'5'' C. 
In column Wh^o is given the corresponding total weight when the hydrometer is 
immersed in distilled water of 19"5°C. The value found for the specific gravity for 
each added weight is given under S. 

'J'he values found for the specific gravity in this series are greater than those found 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 51 



in series (h), (c), or (d), but it will be noticed that the values agree very well inter 
se, the maximum difference, or amplitude, between any two values being only 19 in 
the sixth decimal place. For series b, c, and d the amplitudes are 25, 32, and 46 
respectively ; or if we take the difference between the highest and the lowest value 
of the specific gravity in any of the three series, the amplitude is only 49 in the sixth 
deciinal place. 

Attention should be called to the number of figures after the decimal place given 
in columns t, dwt, W, and Wh^o, which is greater than is consistent with the prob- 
ability of experimental error, but they are inserted in order to facilitate the better 
understanding of the various steps in the process of correcting for temperature, and to 
avoid any ambiguity which might arise were the temperatures only given to two 
places of decimals, and the corresponding weights to only four decimal places. The 
effect of inclusion or exclusion of the last two decimal places in W and Wg^o has 
no appreciable effect upon the value of the sixth decimal figure in the specific gravity. 

Twenty-seven salts— KCl, EbCl, CsCl, NaCl ; KBr, EbBr, CsBr ; KI, Rbl, Csl ; 
KClOg, RbClOg, CsClOg; KBrOg, RbBrOg, CsBrOg ; KIO3, RblOg, CsIOg ; KNO3, 
RbNOg, CsNOg, LiNOg, NaNOg, Sr(N03)2, Ba(N03)2, and Pb(N0g)2— have been dealt 
with in solutions varying in concentration from 1/2 to 1/1024 gram-molecule per 1000 
grams of water, and at temperatures of 15", 19'5°, 23°, and 26° C. 

§ 21. Statistics relating to the Range of Variation of Temperature during a 
Series of Observations. — In the following table statistics of the variations of the 
temperature of the liquid, while a total of 1316 series of observations was made 
with hydrometers Nos. 17 and 21 — namely, 837 with No. 17, and 479 with No. 21 — 
are given. In 68 per cent, of the series made with No. 17 there was no sensible 



A. 


B. 


C. 


A. 


B. 


C. 


Number of Series of 




The Nurabeis 


Number of Series of 




Tlie Numbers 


Observations made 


During which the 


under A expressed 


Observations made 


During which the 


under A expressed 


with Hydrometer No. 


Range of 

Variation of 

Temperature was 


as Percentages. 


with Hydrometer No. 


Range of 

Variation of 

Temperature was 


as Percentages. 


17 


21 


17 


21 


17 


21 


17 


21 






°C. 










°C. 






568 


265 


000 


68-0 


552 


2 








14 


0-2 


00 


48 


51 


0-01 


5-7 


10-6 


3 







15 


0-4 


0-2 


68 


54 


0-02 


8-1 


11-3 


1 







16 


0-12 


0-2 


19 


23 


003 


2-3 


5-0 


1 







17 


0'12 


0-2 


12 


16 


0-04 


1-4 


3-3 










18 


0-00 


0-2 


34 


18 


0-05 


4-0 


3-8 










19 


0-00 


0-2 


7 


8 


0-06 


0-8 


1-7 


6 


2 





20 


0-71 


0-4 


1 


3 


0-07 


012 


0-6 


1 








21 


0-12 


0-0 


13 


10 


0-08 


1-6 


2-1 


1 








24 


0-12 


0-0 


4 


1 


0-09 


0-5 


0-2 





1 





28 


0-00 


0-2 


44 


18 


0-10 


5-2 


3-8 


1 





0-30 


0-12 


0-0 


1 

1 



3 


0-11 
0-12 


0-12 
012 


00 
0-6 




















1 


1 


013 


0-12 


0-2 


Total, 837 


479 




100-00 


100-00 



52 



MR J. Y. BUCHANAN ON THE 



Departure of the Mean Temperature from the Standard Temperature 
during a Series of Observations. 



A. 


B. 


C. 


A. 


B. 


C. 


Number of Series of 

Observations made 

with Hydrometer No. 


During which the 
Departure of the 
Mean Tempera- 
ture from the 
Standard 
Temperature was 


The Numbers 

under A expressed 

as Percentages. 


Number of Series of 

Observations made 

with Hydrometer No. 


During which the 
Departure of the 
Mean Tempera- 
ture from the 
Standard 
Temperature was 


The Numbers 

under A expressed 

as Percentages. 


17 


21 


17 


21 


17 


21 


17 


21 


554 
35 
69 
19 
17 
32 
10 

3 
19 

3 
52 

5 


260 

39 

56 

26 

18 

12 

17 

3 

9 

2 

23 

2 


°G. 
0-000 
0005 
0010 
0015 
0020 
0-025 
0-030 
0-035 
0-040 
0-045 
0-050 
0-055 


66-3 
4-2 
8-2 
2-3 
2-0 
3-8 
1-2 
0-4 
2-3 
0-4 
6-2 
0-6 


54-2 
8-2 

11-7 
5-5 
3-8 
2-5 
3-5 
0-6 
1-9 
0-4 
4-8 
0-4 


2 
1 
1 
2 
1 

3 
1 
6 
2 


3 
3 
1 


4 



1 



•c. 

0-060 
0-065 
0-070 
0-075 
0-080 
0-085 
0090 
0-095 
0-100 
0-120 


0-2 
0-1 
0-1 
0-2 
Oi 
0-0 
0-4 
01 
0-7 
0-2 


0-6 
0-6 
0-2 
0-0 
0-9 
0-0 
0-0 
0-0 
0-2 
0-0 


Total, 837 


479 


100-0 


1000 



variation of temperature, and the same was the case in 55 "2 per cent, of the series 
made with No. 21. If we consider the series made with both hydrometers for which 
the variation of temperature was not greater than 0'05°, the percentages are almost 
identical, namely, 89-5 for No. 17 and 89-2 for No. 21. 

The maximum departure of the mean temperature from the standard, during any 
single series of observations, was 0'12° C. ; the mean departure was 0'0075° C. 

The maximum range of temperature while a series of nine observations was being 
made was 0"30° C. ; the mean range of temperature for the 1316 series of nine observa- 
tions each was 0"018° C. 

These statistics show that the efforts made to secure constancy of temperature 
were successful. 



Seotion IV. — The Control of the Temperature of the Laboratory. 

§ 22. A laboratory is an inhabited room, and in Northern Europe the temperature of 
such apartments lies generally between 12° and 20° C. Consequently we find that a 
large amount of specific gravity work has been done at 15° C. by Gerlach, at 19-5° C. 
by Kremers, and at 17 '5° C. by others ; while the calorimetric work by Julius Thomsen, 
extending over the last half-century, was all done at the temperature 18° C. 

It is always possible to raise the temperature of a laboratory or a dwelling-room, 
but it is not easy to lower it below the atmospheric temperature outside of the house. 
The temperature of the atmosphere in our latitudes rises often above 20° C. in summer, 
and then a temperature as low as 19'5°C. cannot be maintained. Similarly, in cold 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 53 

winters it is difficult to keep even a fairly well-heated room constantly at as high a 
temperature as 19 "5°. 

We see then that, although the temperature 19 "5° 0. may be very suitable and 
be very easily maintained during the greater part of the year, this may not be the case 
under the extreme conditions of raid-winter and mid-summer. It is therefore necessary, 
besides the particular temperature selected as the basis of the mean temperature of 
the locality, to have a higher temperature to meet the case of a warm summer, and a 
lower temperature to meet that of a cold winter. When the heat outside is such that 
the temperature of the air of the laboratory naturally rises above 19 "5°, I use 23° C. 
as the particular temperature at which all my observations are made. On the rare 
occasions when this temperature is too low, I use 26° C, and this is a very useful 
temperature in tropical countries. At sea, even in equatorial regions, the highest 
particular temperature that would be required is 30° C. In winter the temperature 
of a laboratory or other inhabited room should never fall below 12° C. In these 
circumstances 15° C. has been adopted as the particular temperature. 

The room used as laboratory should be of moderate dimensions, rather small than 
large, because it must be occupied only by the experimenter, and he must have absolute 
control over it. It should be illuminated by sky light, and the direct rays of the sun 
must be absolutely excluded. In our latitudes this means that the window must have 
a northern exposure. The room should be furnished with central heating, preferably 
by hot water. 

When the particular temperature required for the liquid, the specific gravity of 
which is to be determined, is 19 '5° C, the temperature of the air should be maintained 
at 19*0° to 19 "3°, but this depends on the room. When the temperature of the liquid 
has been brought exactly to 19 '5°, it will remain constant at this temperature when the 
temperature of the air in the neighbourhood of the cylinder is at 19 '2° or 19 "3°, because 
the heat which is removed from it by convection is equal to that supplied to it by 
radiation, principally from the experimenter himself. 

The constant temperature of the liquid in the cylinder is the integral efi"ect of a 
number of separate elements, of which the principal are the temperature of the air 
within and without the room and that of the experimenter. In mild weather, when 
the temperature of the air outside is about 14° to 17° C, it is generally very easy to 
regulate the temperature of the air in the laboratory so that that of the liquid in the 
cylinder may remain constantly at 19*5° for the duration of the experiment. In cold 
weather, however, when the temperature of the air outside may be 8° or 10° or even 
more below the temperature of the room, the liquid is apt to experience sensible cooling 
by radiation to the outside through the glass of the window. (Of course the window 
must always be shut when experiments are being made, because the slightest 
draught striking the exposed stem of the hydrometer disturbs the reading as well as 
the temperature.) In these circumstances it is often difficult to maintain its tempera- 
ture at 15° for the duration of the experiment. 



54 MR J. Y. BUCHANAN ON THE 

§ 23. It was by an accident, which for the moment was annoying, that I found a very 
simple and efficacious means of counteracting this loss of heat on the part of the liquid. 

At Edinburgh, in December 1902, I was making some observations on the specific 
gravity of various saline solutions, and had got the temperature of the laboratory so 
that 1 had no difficulty in carrying out the determinations without change of tempera- 
ture of the liquid. In latitude 56° in December, even in clear weather, night begins 
to fall early in the afternoon, and, in the middle of the work, I had to light the gas 
in order to be able to continue it. The gas jet was fully a metre above the cylinder, 
and the light which it gave, though sufficient, was far from brilliant. When I had 
finished the series of the usual number, nine individual observations, and removed the 
hydrometer, I took the temperature of the liquid, expecting to find that it had remained 
sensibly constant as before ; but, instead of this, it showed a rise of several tenths of 
a degree. As the gas had been burning for only a few minutes, it was impossible for 
it, in the time, to raise the temperature of the air of the laboratory so that it, in its 
turn, could raise that of the liquid in the cylijider by any sensible amount ; and, in 
fact, on regarding the thermometer used for indicating the temperature of the air, it 
was found that this had remained unchanged. For the moment I was perplexed ; the 
shortness of the interval between the lighting of the gas jet and the production of the 
heat-effect on the liquid in the cylinder situated at a considerable distance below it 
puzzled me. However, after some reflection / 'perceived that it must he an effect of 
radiation from the luminous flame of the gas jet, which, as soon as it was lighted, 
dispatched its heat-rays in all directions and ivith the velocity of light. It was this 
radiation that penetrated the cylinder and raised the temperature of the liquid in so 
short a time ; and this eff"ect could be produced by no other agency. 

Further reflection showed me that the agency which can raise the temperature of 
the liquid in this way can also prevent it falling, I gave eff"ect to this idea by attaching 
a luminous gas lamp by an india-rubber tube to one of the gas cocks on the working 
bench of the laboratory, and placed it between the cylinder and the window. The 
gas burner was at a height above the table inferior to that of the top of the cylinder, 
and the standard which carried it could be shifted towards or away from the cylinder 
at will. 

It must be remembered that the provision of this gas flame was not for the purpose 
of heating the air and raising or maintaining its temperature up to a certain degree ; it 
was to supply the cylinder and liquid with heat, and that without warming the inter- 
vening air ; and the heat so communicated from a distance was to be regulated so as 
exactly to make good that which they were dissipating. 

As the combustion of the gas necessarily heated air which went upwards, it was 
important to secure the supply of radiant heat with the least possible combustion of 
gas. This was effected by reducing the flame so as to give the smallest possible flame 
which was sufficiently luminous to furnish the necessary radiation at a distance which 
was found to be convenient, generally from 50 to 75 centimetres. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 55 



§ 24. The room in which the work is carried out is 16x8 by 10 feet high, the 
16-feet length being almost due north and south. The room is lighted by a fairly large 
window on the north wall, and there is a door in the south wall. There is a small fume 
chest projecting half way along the east wall, closed in by a casement window. 

Against the west wall is a strong, well-made table about 48 x 27 inches, whereat 
all the observations are made. 

The room is fitted with sfas and electric liffht, and a bench runs the whole width 
of the room under the window, and is fitted with gas fittings. The room is warmed 
by a steam radiator. The following plan gives an idea of the general disposition of 
the room : — 



Window recess, 

same level as 

bench. 



Sink. 



Bench. 
X Compensating luminous flame. 



O 03 
0> ■ 







•-d 












hrj\^ 




S. 


B 


> 


o 


f 




'^ y/ 




1 







o- 



-Radiator. 




For a typical working day the 6th December 1911 has been selected. 

On arriving at the laboratory about 10.0 a.m. the room temperature was 17'6'C., 
although the radiator was working at full pressure. 

The meteorological conditions, such as pressure and relative humidity, were noted 
and recorded for the purpose of reducing weights "to vacuo" where solutions were to 
be prepared. 

The temperature of the air by this time was about 18-5° C, and the day was cold. 
A bunsen was lit in the fume chest, which is at the back of the experimenter when 



56 MR J. Y. BUCHANAN ON THE 

seated at the table making observations. The door of the fume chest being closed, the 
heat from the bunsen is very evenly distributed into the room. 

The two hydrometers which were used in the experiments were taken from their 
cases, were each immersed in a small cylinder containing distilled water at about 1 9 '6'' C, 
and left in to attain a temperature about 19 50° C, a temperature at which the specific 
gravity observations were to be made. 

The solution used on this occasion was a ^ gram-molecule solution of the potassium 
chloride and iodide mixed in equimolecular proportions, the molecular weight assigned 
being the mean between the molecular weight of potassium chloride and that of 
potassium iodide. The weight of salt represented by y^ gram-molecule was dissolved 
in 1000 grams of water and was prepared overnight. 

The bottle of solution had been standing near the radiator for some time to attain 
a temperature near to 19 50° C. 

The solution was now poured into the cylinder used for the experiments, and the 
quantity was such that when the largest hydrometer was immersed to its fullest extent 
the surface of the solution was fully an inch below the rim of the cylinder, a precaution 
which obviates difficulties in reading likely to be occasioned by irregularities in the 
glass occurring near the top. 

The cylinder containing the solution was then placed on the table at a convenient 
altitude for making observations (the foot of the cylinder resting on a thickness 
of sixteen folds of soft German filter paper to form a non-conducting surface), and 
the temperature, as observed with a standard thermometer divided into yV^li^ inch, 
each division being of such a size as to enable one to read y^^jths of a degree in 
temperature with comparative ease, was 1870° C, the air temperature by this time- 
being 19-0° C. 

An expeditious and effective method for rapidly raising the temperature of the 
cylinder and contents, by stroking the side of the cylinder with the palms of the hands, 
was adopted, and by this means the temperature was quickly raised to 19'50°C. exactly. 

The time was then 10.50 a.m., and the room temperature 19' 1 C, so the bunsen 
was lowered somewhat, and the radiator turned to half way. 

On removing the hydrometers from their respective cylinders and drying them, the 
temperature of the water in which they had been immersed was 19"35° C. in both cases, 
so that the hydrometers were presumably at that temperature. 

The temperature of the solution was still at 19*50° C, and the air temperature 
19*20° C, so that the conditions were suitable for commencing observations. 

After removing the thermometer from the solution and immersing it in one 
of the cylinders of distilled water, the hydrometer No. 17 was taken from its case 
and gently lowered into the solution, and an initial added weight placed on the 
top of the stem of the hydrometer, the time of commencement of the experiment 
being noted. 

Nine successive readings, as the results of addition of nine weights, eight of them 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 57 

havino; the value of 0*1 gram each, were taken, and after the ninth observation the 
thermometer was taken out of the distilled water and dried ; the hydrometer was 
removed from the experimental solution and put into the cylinder containing the 
distilled water, and the temperature of the experimental solution observed by immersing 
the thermometer and gently stirring the solution. 

The hydrometer was then dried and replaced in its case. 

To complete the experimental data, the air temperature and time of completion of 
the experiment were noted. 

The following is the record of the experiment : — 

Y^ gram-molecule solution K . 

Hydrometer 17. 
A.ir temperature . . = 19 '20° C. 

Initial solution temperature = 19'50° C. 
Time . . . . =11.3 a.m. 



Added "Weight. 
r32 grams. 




Reading. 
5*2 millimetres. 


1*42 „ 




16-3 


>> 


1-52 „ 




27-5 


j> 


1-62 „ 




38-4 


>> 


1-72 „ 




49-4 


)> 


1-82 „ 




59-2 


)> 


1-92 „ 




70-1 


)> 


2-02 „ 




82-1 


5> 


2-12 „ 




93-0 


)) 


Mean added weight 


. 


= 172 gram. 


Mean reading 


. 


. 


= 49*02 millimetres 


Air temperature 


. 


= 19-30° C. 


Final solution 


temperature 


= 19-50° C. 


Time 


. 




= 11.16 a.m. 



It will be seen that the air temperature rose 0-1° C. during the experiment, 
which occupied 1 3 minutes, and the solution temperature remained constant ; so the 
bunsen in the fume chest was lowered to a further extent, and then preparations 
for the next experiment, conducted in precisely the same manner with hydrometer 
No. 3, were made. 

As this section deals only with temperature conditions, the following table has been 
drawn up to show the temperature conditions which prevailed during the experiments 
which were made on the -^^ and ^ gram-molecule solution of the potassium salt of 
the mixed halides (chloride and iodide), these being the two solutions experimented 
upon during the day : — 

TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 8 



58 



MR J. Y. BUCHANAN ON THE 







Quantity of Salt 
















Hydro- 
meter 
used. 


Number 
of Experi- 
ment. 


in 1000 grams 

of Water, 

expressed in 

gram-molecules. 

m. 


Initial Air 
Tempera- 
ture. 


Initial 
Solution 

Tempera- 
ture. 


Time of 
Commence- 
ment of 
Experiment. 


Final Air 
Tempera- 
ture. 


Final 
Solution 
Tempera- 
ture. 


Time of Com- 
pletion of 
Experiment. 


Duration 
of Experi- 
ment in 
M-inutes. 








°C. 


°C. 




°C. 


"C. 






No. 17 


1 


1 


19-20 


19-50 


11.3 a.m. 


19-30 


19-50 


11.16 a.m. 


13 


„ 3 


2 


J) 


19-30 


19-50 


11.20 „ 


19-30 


19-50 


11.34 „ 


14 


» 17 


3 


fy 


19-30 


19-50 


11.40 „ 


19-30 


19-50 


11.54 „ 


14 


„ 3 


4 


)j 


19-30 


19-50 


11.59 „ 


19-30 


19-50 


12.12 p.m. 


13 


„ 17 


5 


}) 


19-30 


19-50 


12.17 p.m. 


19-30 


19-50 


12.31 „ 


14 


„ 3 


6 


)) 


19-30 


19-50 


12.36 „ 


19-30 


19-50 


12.50 „ 


14 


No. 17 


7 


1 


19-30 


19-50 


1.45 p.m. 


19-30 


19-50 


1.57 p.m. 


12 


„ 3 


8 


)J 


19-30 


19-50 


2.5 „ 


19-30 


19-50 


2.17 „ 


12 


» 17 


9 


J) 


19-30 


19-50 


2.22 „ 


19-30 


19-50 


2.34 „ 


12 


,, 3 


10 


)J 


19-35 


19-50 


2.45 „ 


19-35 


19-50 


2.59 „ 


14 


„ 17 


11 


)? 


19-30 


19-50 


3.4 „ 


19-35 


19-50 


3.16 „ 


12 


„ 3 


12 


J) 


19-30 


19-50 


3.22 „ 


19-30 


19-50 


3.35 „ 


13 


„ 17 


13 


)! 


19-30 


19-50 


3.42 „ 


19-30 


19-50 


3.56 „ 


14 



It will be seen that the initial and final solution temperatures were constant to 
within 0'01° C. throughout the series of experiments. There were slight variations in 
the air temperature of the room, the widest range being from 19"20° C. to 19"35° C. 
The rise was occasioned by turning on the radiator full for a few minutes and opening 
the door for fresh air, but no change occurred in solution temperature, so that latitude 
can be given in the range of air temperatures ; but from experience it is not advisable 
to go below 19 '2° C. unless direct radiation can be supplied to the solution, as shown in 
the earlier part of this section, the source of which can be effectively controlled. 

In the conditions which obtain in this laboratory, it is possible to conduct a series 
of experiments extending over the day and to maintain the temperature of each 
solution constant for at least fourteen minutes if the temperature of the air is kept 
0*3° lower than that of the solution. 

§ 25. While the conditions which have been described are all essential for complete 
success in hydrometric work from the point of view of constant temperature, it may, 
and does occasionally, happen that, even after adopting all the precautions mentioned 
above, a series of observations will be taken, and then the solution temperature will be 
found to have changed, and with this change there has been a deviation in the value of 
specific gravity, certainly in the most extreme case amounting to only a few units in 
the 5th decimal place ; but the deviation coupled with the temperature change has, 
in the most recent work, justified its elimination from the remaining series of observa- 
tions which are perfect, in that the results of the other series agree iriter se and no 
change in solution temperature has occurred. 

An example of such an occurrence happened on 4th April 1911, when a series of 
hydrometric observations was made upon a solution containing ^ gram-molecule NaCl 
in 1000 grams water. It was the first series of the day, and although the initial and 
final air temperatures for this experiment were both 19 "30° C, the solution temperature 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 59 

changed from 19'50° C. at the commencement of the experiment to 19"41° C. when the 
observations were completed. The value of the specific gravity as calculated from this 
experiment was r004542. This value was not included in the accepted results, and 
the second series of experiments with the same hydrometer, where no change of tempera- 
ture occurred in the solution during the experiment, gave a specific gravity value of 
1'004570, while the mean of the whole series was 1 004579. 

This variation in the solution temperature could only be due to the temperature of 
the hydrometer itself being considerably lower than that of the solution, and this factor 
operates in the twofold manner of lowering the solution temperature, and by virtue of 
the fact that there is a contraction in the volume of the hydrometer at a lower tempera- 
ture, the added weight to sink it to a given scale division is less than at the higher 
temperature, so that the specific gravity value is lower than that obtained when the 
hydrometer is at the standard temperature. This is indubitably the explanation of the 
change in the solution temperature during the experiment quoted above, and may not 
improbably account for the observation that the first reading of the day is sometimes 
not comparable with the later results obtained in observations made on the same 
solution. 

It is not possible to directly ascertain the temperature of the hydrometer, and since 
it is necessary that it should be acclimatised to the experimental temperature, the 
precaution of immersing the hydrometers in distilled water at the expey'imental 
temperature for some tim,e before commencing hydrometric observation is im^portant. 
It ensures that the hydrometer shall be at the experimental temperature and that its 
volume shall be normal for the given temperature. 

The water value of one of the hydrometers is 11 gram-degrees centigrade. It is 
possible to calculate the temperature of the hydrometer which would reduce the 
solution temperature from 19'50°C. to 19*41° C. in the instance mentioned above, 
assuming no loss of heat due to radiation (this loss is negligible in any case under 
the conditions of experiment). The weight of the water in solution is about 600 grams 
(specific heat= l). Hence, applying the principle of the determination of specific heat 
by the method of mixtures, the temperature of the hydrometer to produce this must 
be 14-59° C. 

The effect of immersing the hydrometer at that temperature was to make the earlier 
readings lower than they would have been if the hydrometer were at normal temperature ; 
but the hydrometer is expanding, because it is in a warmer medium, so that the later 
readings in the same series of observations would approach the values that would have 
been obtained if the hydrometer had been at normal temperature. 

It is not difficult to show that the hydrometer was at this temperature (14'59° C.) 
at the beginning of the experiment, since we have its coefficient of expansion, 
namely^, 0*003 c.c. per degree difference of temperature. 

If we assume a mean hydrometer temperature of 17'00° C. — which is the mean 
between the initial and final hydrometer temperatures — then the volume of solution 



60 MR J. Y. BUCHANAN ON THE 

not displaced on account of the shrinkage in bulk of the hydrometer due to lower 
temperature = 0*003 x 2"5 = 0"0075 c.c, and its weight is "007 5 3 gram (specific gravity 
of solution being 1 "004542). 

Hence, in correcting this value of specific gravity, the weight of solution displaced 
by the hydrometer is 182-53030 grams + 0-00753 gram= i82'53783 grams. The weight 
of distilled water displaced = 1 8r70496 grams. The specific gravity is therefore 
r004583, which closely agrees with the mean specific gravity of 1-004579. 

The difference of scale reading occasioned by this difference due to temperature, and 
which is equivalent to an added weight of 0-00753 gram, is 0-82 mm., and is quite 
an appreciable quantity when it affects all the readings in a series to this extent. 

The importance, therefore, of the precaution for commencing the series of experi- 
ments with the hydrometers at the standard temperature by the simple device of 
immersing them in water at, or a little above, the standard temperature for some time, 
and drying them before commencing the experiments, is at once apparent. 

In the following section the experimental work is embodied in a number of tables. 
Full explanation is given for each class of tables. Supplemental work which was done 
during the preparation of this memoir is described and discussed in later sections. 



[Tables. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 



61 



Section V.— TABLES. 

26. A. General Tables, giving the Facts of Observation ; namely, W, the weight, in grams, of 
the solution which contains m gram-molecules of the salt dissolved in 1000 grams of water ; 
S, the specific gravity of this solution at the temperature, T, referred to that of distilled 
water at the same temperature as unity. From these data. A, the displacement of the 
solution, is obtained by the equation A = W/S, and it is expressed in grams of water having 
the temperature T. The symbol adopted for this unit is Gx- The numbers printed in 
italics refer to specific gravities observed at the temperature printed in italics. 

TRIAD OF CHLORIDES. 

Table No. 1. 

POTASSIUM CHLORIDE. KCl = 74-6. 

T= 19-5° and ^5<?° C. 





Weight of 
Solution. 


Specific 


Displacement. 


Differences of 


Differences of 
Lo"^arithms of 




Grams. 


Giavity. 


Gt. 


Displacements. 


Displacements. 


m. 


W. 


S. 


W/S = A. 


dl^. 


d log A. 


1/2 


1037-3000 


1022977 


1014001 




(3-0060385)* 


1/4 


1018-6500 


1-011670 


1006-899 


7102 


0-0030522 


1/8 


1009-3250 


1-005889 


1003-416 


3-483 


0-0015052 


1/16 


1004-6625 


1-002973 


1001-684 


1-732 


0-0007500 


1/32 


1002-3312 


1-001489 


1000-841 


0-843 


0-0003658 


1/64 


1001-1656 


1-000741 


1000-423 


0-418 


0-0001811 


1/128 


1000-5828 


1-000365 


1000-217 


0-206 


0-0000895 


1/256 


1000-2914 


1-000193 


1000-098 


0-119 


00000517 


1/512 


1000-1457 


1-000082 


1000-064 


0-034 


0-0000149 


1/16 


1004.-6625 


1-00292 J^ 


1001-7 SS 










Tablb No. 2. 










RUBIDIUM CHLORIDE. RbCl = 1 


21-0. 








T= 19-5° and ^^-O" C. 






1/2 


1060-5000 


1-043144 


1016-637 




(3-0071662) 


1/4 


1030-2500 


1-021868 


1008-202 


8-435 


0-0036185 


1/8 


1015-1250 


1-011023 


1004-057 


4-145 


0-0017892 


1/16 


1007-5625 


1-005531 


1002-020 


2037 


0-0008819 


1/32 


1003-7812 


1-002772 


1001-006 


1-014 


0-0004397 


1/64 


1001-8906 


1-001400 


1000 489 


0-517 


00002241 


1/128 


1000-9453 


1-000707 


1000-238 


0-251 


0-0001090 


1/256 


1000-4726 


1-000330 


1000-122 


0-116 


0-0000502 


1/512 


1000-2363 


1-000163 


1000-073 


0-049 


0-0000215 


1/16 


1007-6625 


1-005485 


1002 066 










Table No. 3. 










CiESIUM CHLORIDE. CsCl = 16{ 


?-5. 








T= 19-5° and ^5-6>° C. 






1/2 


1084-2500 


1-062572 


1020-401 




(3-0087709) 


1/4 


1042-1250 


1-031739 


1010-066 


10-335 


0-0044198 


1/8 


1021-0625 


1-015994 


1004-989 


5-077 


0-0021887 


1/16 


1010-5312 


1-008036 


1002-475 


2-514 


0-0010875 


1/32 


1005-26.^6 


1-004035 


1001-225 


1-250 


00005417 


1/64 


1002-6328 


1-002027 


1000-604 


0-621 


0-0002694 


1/128 


1001-3164 


1-001025 


1000-291 


0-313 


0-0001362 


1/256 


1000-6582 


1-000514 


1000-144 


0-147 


0-0000636 


1/512 


1000-3291 


1-000249 


1000079 


0-065 


0-0000280 


1/16 


1010-5812 


1007954 


1002-557 







* The entry in brackets in the column d log A gives log A for the solution of highest concentration in the series. With 
it and the following values of d log A in the same column the values of log A for each solution in the series can be calculated. 



62 



MR J. Y. BUCHANAN ON THE 



A. General Tables, giving the Facts of Observation in the columns 

under m, W, and S. 

TRIAD OF BROMIDES. 



Table No. 4. 
POTASSIUM BROMIDE. KBr= 119-1. 
T = 19-5° and ^50° C. 





Weight of 
Solution. 

Grams. 


Specific 
Gravity. 


Disitlacement. 


Differences of 
DispUcemeuts. 


Differences of 
Logarithms of 
Displacements. 


m: 


Wt 


- S. 


W/S = A. 


dt.. 


d log A. 


1/2 


1059-5500 


1-041278 


1017-547 




(3-0075546) 


1/4 


1029-7750 


1-020903 


1008-690 


8-857 


0-0037967 


1/8 


1014-8875 


1-010528 


1004314 


4-376 


0-0018882 


1/16 


1007-4437 


1-005279 


1002153 


2161 


0-0009346 


1/32 


1003-7218 


1-002638 


1001081 


1-072 


0-0004648 


1/64 


1001-8609 


1-001306 


1000-554 


0-527 


0-0002287 


1/128 


1000-9304 


1-000652 


1000-278 


0276 


00001195 


1/256 


1000-4652 


1-000325 


1000-139 


0-139 


00000602 


1/512 


1000-2326 


1 000158 


1000074 


0-065 


0-0000283 


1/16 


1007-US7 


1-005306 


1002-126 










Table No. 5. 











RUBIDIUM BROMIDE. RbBr = 165-5. " 


- - --' - - 






T = 19-5°C. 






1/2 


1082-7500 


1-061247 


1020-262 




(3-0087116} 


1/4 


1041-3750 


1-031081 


1009-983 


10-279 


0-0043972 


1/8 


1020 6875 


1-015669 


1004-941 


5-042 


0-0021737" 


1/16 


1010-3437 


1-007868 


1002-45G 


2-485 


0-0010751 


1/32 


1005-1718 


1-003945 


1001-222 


1-234 


00005350 


1/64 


1002-5859 


1001957 


1000-627 


0-595 


0-0002578 


1/128 


1001-2929 


1-000984 


1000.308 


0-319 


00U01385 


1/256 


1000-6464 


1-000457 


1000-189 


0-119 


0-0000518 


1/512 


1000 3232 


1000233 


1000-090 


0-099 


0-0000430 


1/1024 


10001616 


1-000079 


1000-082 


0-008 


0-0000033 


- — 


- -•- - -•■ 


Table No. 6. 

CAESIUM BROMIDE. CsBr = 213 
T=19-5°C. 


•0. 


- .1 


""112 


1106-5000 


1-080935 


1023-650 






(3-0101517). 


1/4 


1053-2500 


1-041011 


1011-756 


11-894 


0-0050756 


1/8 


1026-6250 


1-020702 


1005-802 


5-954 


0-0025632 


1/16 


1013-3125 


1-010409 


1002-873 


2-929 


0-0012666 


1/32 


1006-6562 


1-005182 


1001-466 


1-407 


0-0006097 


1/64 


1003-3281 


1-002631 


1000-695 


0-771 


0-0003346 


1/128 


1001-6640 


1-001270 


1000-393 


0-302 


00001309 


1/256 


1000-8320 


1000607 


1000-224 


0-169 


0-0000732 


1/512 


1000-4160 


1-000308 


1000-108 


0-116 


00000507 


1/1024 


1000-2080 


1000145 


1000 063 


0-045 


0-0000195 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 



63 



A. General Tables, giving the Facts of Observation in the columns 

under m, W, and S. 

TRIAD OF IODIDE&. 

Table No. 7. 
POTASSIUM IODIDE. KI = 166-1. 







T = 


19-5° C. 








Weight of 
Solution. 

Grams. 


Specific 
Gravity. 


Displacement. 

Gt. 


Differeuces-of — 
Displacements. 


Differences of 


Logarithms of 
Displacements. 


m. 
1 

3/4 


1166-1000 

1124-5750 


- -S.- 

1-114617 
1-087124 


W/S=A. -- 

1046-189 
1034-449 


11-740 


^logA. 
(3-0196102) 
0-0049008 


1/2 


1083-0500 


1-058929 


1022-778 


11-671 


00049276 . 


1/4 


1041-5250 


1-029906 


1011-281 


11-497 


0-0049095 


1/8 


1020-7625 


1-015104 


1005-574 


5-707 


0-0024579 


1/16 


1010-3812 


1-007588 


1002-772 


2-802 


0-0012119 


1/32 


1005-1906 


1-003790 


1001-395 


1-377 


00005967 


1/64 


1002-5953 


1-001899 


1000-695 


0-700 


0-0003038 


1/128 


1001-2976 


1-000950 


1000-347 


0-348 


00001509 


1/256 


1000-6488 


1-000480 


1000168 


0-179 


0-0000775 


1/512 -^ 


- -lOOa-3244- . . 


.1-000235 


1000-089 


079 - 


0-0000345 


1/1024 


1000-1622 


1-000122 


1000-040 


0-049 


0-0000212 






Table No. 8. 










RUBIDIUM IODIDE. Rbl = 212 


-5. 








T=19-5°C. 






1/2 
1/4 


1106-2500 
1053-1250 


1-078421 
1-039778 


1025-805 
1012-836 


12-969 


(3-0110649) 
0-0055256 


1/8 


1026-5625 


1-020010 


1006-424 


6-412 


0-0027583 


1/16 


1013-2812 


1-010046 


1003-203 


3-221 


0-0U13921 


1/32 


1006-6406 


1-005030 


1001-602 


1-601 


0-0006934 


1/64 


1003-3203 


1-002505 


1000-813 


0-789 


0-00U3423 


1/128 


1001-6601 


1-001237 


1000-422 


0-391 


0-0001695 


1/256 


1000-8300 


1-000612 


lOOu-218 


0204 


0-0000888 


1/512 


1000-4150 


1-000272 


1000-143 


0075 


0000325 


1/1024 


1000-2075 


1-000146 


1000-061 


0-082 


0-0000353 






Table No. 9. 










CESIUM IODIDE. Csl = 260-C 


. 








T=19-5°C. 






1/2 
1/4 


1130-0000 
1065-0000 


1-097427 
1-049480 


1029-681 
1014-788 


14-893 


(3-0127028) 
0-0063273 


1/8 


1032-5000 


1-024973 


1007-343 


7-445 


0-0031978 


1/lS 


1016-2500 


1-012529 


1003-675 


3-668 


00015845 


1/32 


10081250 


1-006299 


1001-814 


1-861 


0-0008057 


1/64 


1004-0625 


1-003120 


1000-939 


0-875 


00003795 


1/128 


i 1002-0312 


1-001546 


1000-484 


0-455 


00001975 


1/256 


1 1001-0156 


1-000738 


1000-277 


0-207 


0-0000S98 


1/512 


1000-5078 


1-000272 


1000-235 


0-042 


00000181 


1/102'J, 


1 1000-2539 


1-000100 


1000-153 


0-082 


0-0000355 



64 



MR J. Y. BUCHANAN ON THE 



A. General Tables, giving the Facts of Observation in the columns 

under m, W, and S. 



TKIAD OF IODIDES. 

Table No. 10. 

POTASSIUM IODIDE. KI = 166-1. 

T = 23-0° C. 





Weight of 
Solution. 

Grams. 


Specific 
Gravity. 


Displacement. 

Gt. 


Differences of 
Displacements. 


Differences of 
Logarithms of 
Displacements. 


m. 
1/2 
1/4 
1/8 
1/16 
1/32 
1/64 
1/128 
1/256 


W. 
1083-0500 
1041-5250 
1020-7625 
1010-3812 
1005-1906 
1002-5953 
1001-2976 
1000-6488 


S. 
1-058639 
1-029717 
1-014990 
1-007544 
1-003761 
1-001919 
1-000950 
1-000497 


W/S = A. 

1023-058 
1011-467 
1005-687 
1002-816 
1001-424 
1000-675 
1000-347 
1000-152 


dA. 

11-591 
5-780 
2-871 
1-392 
0-749 
0-328 
0-195 


d log A. 
(3-0099005) 
0-0049486 
0-0024888 
0-0012416 
0-0006034 
0-0003249 
0-0001420 
0-0000849 


Table No. 11. 

RUBIDIUM IODIDE. Rbl = 212-5. 

T = 23-0''C. 


1/8 

1/16 

1/32 

1/64 

1/128 

1/256 


1026-5625 
1013-2812 
1006-6406 
1003-3203 
1001-6601 
1000-8300 


1-020075 
1-010092 
1-005043 
1-002555 
1-001277 
1-000653 


1006-360 
1003-157 
1001-589 
1000-763 
1000-382 
1000-176 


3-203 

1-568 
0-826 
0-381 
0-206 


(3-0027534) 
0-0013943 
0-0006793 
0-0003582 
0-0001653 
0-0000893 


Table No. 12. 

CiESIUM IODIDE. Csl = 260-0. 
T = 23-0°and^e-0°C. 


1/8 

1/16 

1/32 

1/64 

1/128 

1/25G 

1116 


1032-5000 
1016-2500 
1008-1250 
1004-0625 
100 J -03 12 
1001-0156 

1016-2500 


1-025081 
1-012637 
1-006341 
1-003163 
1-001596 
1-000814 

1-01262S 


1007-237 
1003-608 
1001-772 
1000-896 
1000-434 
1000-201 

1003-582 


3-629 
1-836 
0-876 
0-462 
0-233 


(3-0031318) 
0-0015675 
0-0007950 
0-0003799 
0-0002005 
0-0001010 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 



65 



A. General Tables, giving the Facts of Observation in the columns 

under m, AV, and S. 

TRIAD OF NITRATES. 



Table No. 13. 

POTASSIUM NITRATE. KNO3 = 101-1. 

T=19-5°C. 





Weight of 
Solution. 


Specific 


Displacement. 


Differences of 


Differences of 




Grams. 


Gravity. 


Gt. 


Displacements. 


Displacements. 


OT. 


W. 


S. 


W/S = A. 


d^. 


d log A. 


1/2 


1050-5500 


1-030564 


1019-410 




(3-0083492) 


V-t 


1025-2750 


1-015533 


1009-593 


9-817 


0-0042028 


1/8 


1012-6375 


1-007873 


1004-727 


4-866 


0-0020983 


1/lG 


1006-3187 


1-003968 


1002-342 


2-385 


00010319 


1/32 


1003-1593 


1-002013 


1001-144 


1-198 


00005194 


1/64 


1001-5796 


1001004 


1000-575 


0-569 


00002463 


1/128 


1000-7898 


1-000509 


1000-281 


0-294 


0-0001283 







Tabi 


.E No. 14. 










RUBIDIUM NITRATE. RbN03 = 147-5. 








T=19-5° C. 






1/2 


1073-7500 


1-050634 


1022002 




(3-0094517) 


1/4 


1036-8750 


1-025698 


1010-897 


11105 


0-0047448 


1/8 


1018-4375 


1-012973 


1005-394 


5-503 


0-0023704 


1/16 


1009-2188 


1-006597 


1002-604 


2-790 


0-0012068 


1/32 


1004-6094 


1-003355 


1001-250 


1-354 


0-0005870 


1/64 


1002-3047 


1-001750 


1000-553 


0-697 


0-0003024 


1/128 


1001-1523 


1-000920 


1000-232 


0-321 


0-0001393 


1/256 


1000-5762 


1-000458 


1000-118 


0-114 


0-0000495 






Table No. 15. 

CESIUM NITRATE. CsN03 = 19J 
T=19-5° C. 


)-0. 




1/4 


1048-7500 


1-035619 


1012-679 




(3-0054720) 


1/8 


1024-3750 


1017961 


1006-301 


6-378 


0-0027440 


1/16 


1012-1875 


1-009041 


1003-118 


3-183 


0-0013757 


1/32 


1006-0937 


1-004585 


1001-501 


1-617 


0-0007006 


1/64 


1003-0468 


1-002247 


1000-798 


0-703 


0-0003049 


1/128 


1001-5234 


1-001146 


1000-376 


0-422 


0-0001830 


1/256 


1000-7617 


1-000604 


1000-157 


0-219 


0-0000952 



TRANS. ROY. SOC. EDIN., VOL, XLIX. PART I. (NO. 1). 



66 



MR J. Y. BUCHANAN ON THE 



A. General Tables, giving the Facts of Observation in the columns" 

under m, W, and S. 



TRIAD OF CHLORATES 

Table No. 16. 
POTASSIUM CHLORATE. 

T = 19-5° and 23-0° C 



KC103=122-G. 





Weight of 
Solution. 

Grams. 


Specific 
Gravity. 


Displacement. 

Gt. 


Differences of 
Displacements. 


Differences of 
Logarithms of 
Displacements. 


m. 


W. 


S. 


W/S = A. 


f?A. 


d log A. 


1/4 


1030-6500 


1-019081 


1011-352 




(3-0049025) 


1/8 


1015-3250 


1-009638 


1005-632 


5-720 


0-0024631 


1/16 


1007-6625 


1-004863 


1002-785 


2-847 


0-0012311 


1/32 


1003-8312 


1-002490 


1001-337 


1-448 


0-0006276 


1/64 


1001-9156 


1-001253 


1000-661 


0-676 


0-0002933 


1/128 


1000-9578 


1-000633 


1000-324 


0-337 


0-0001463 


1/256 


1000-4789 


1-000320 


1000-158 


0-166 


0-0000719 


1/512 


1000-2394 


1-000182 


1000-057 


0-101 


0-0000440 


ijie 


1007-6625 


1-004759 


1002-889 










Table No. 17. 








RUBIDIUM CHLORATE. RbC103 = 


169-0. 








T = 19-5° and 23-0'' (\ 






\/4: 


1042-2500 


1-029153 


1012-726 




(3-0054919) 


1/8 


1021-1250 


1-014679 


1006-353 


6-373 


0-0027417 


1/16 


1010-5625 


1-007356 


1003-183 


3-170 


0-0013700 


1/32 


1005-2813 


1-003691 


1001-584 


1-599 


0-0006926 


1/64 


1002-6406 


1-001863 


1000-775 


0-809 


0-0003406 


1/128 


1001-3203 


1-000919 


1000-400 


0-375 


0-0001626 


1/256 


1000-6602 


1-000459 


1000-200 


0-200 


0-0000867 


1/512 


1000-3301 


1-000218 


1000-111 


0-089 


0-0000386 


1/16 


1010-6625 


1-007331 


1003-20J^ 










Table No. 18. 










CESIUM CHLORATE. CsCI03 = 2 


16-5. 








T=19-5° and 23-0'' G. 






1/4 


1054-1250 


1-039043 


1014-515 




(3-0062587) 


1/8 


1027-0625 


1-019686 


1007-233 


7-282 


0-0031285 


1/16 


1013-5312 


1-009825 


1003-669 


3-564 


0-0015394 


1/32 


1006-7656 


1-004953 


1001-804 


1-865 


0-0008066 


1/64 


1003-3828 


1-002409 


1000-971 


0-833 


0-0003624 


1/128 


1001-6914 


1-001216 


1000-476 


0-495 


0-0002149 


1/256 


1000-8457 


1-000552 


1000-292 


0-184 


0-0000795 


1/512 


1000-4228 


1-000210 


1000-212 


0-080 


0-0000348 


1/16 


1013-5312 


1-009886 


1003-609 







SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 67 

A. General Tables, giving the Facts of Observation in the columns 

under on, W, and S. 

TRIAD OF BROMATES. 



Table No. 19. 

POTASSIUM BROMATE. KBr03 = 167-1. 

T=19-5° and 23'0° C. 



Table No. 21. 
CESIUM BROMATE 

T= 19-5° and ^5-0° C 



CsBr03 = 261-0. 





Weight of 
Solution. 

Grams. 


Specific 
Gravity. 


Displacement. 

Gt. 


Differences of 
Displacements. 


Differences of 
Logarithms of 
Displacements. 


VI. 


W. 


S. 


W/S = A. 


f^A. 


d log A. 


1/4 


1041-7750 


1-030144 


1011-290 




(3-0048760) 


1/8 


1020-8875 


1-015227 


1005-576 


5-714 


0-0024613 


1/16 


1010-4438 


1-007662 1002-761 


2-815 


0-0012173 


1/32 


1005-2219 


1-003846 


1001-370 


1-391 


0-0006026 


1/64 


1002-6109 


1-001921 


1000-688 


0-6S2 


0-0002959 


1/128 


1001-3055 


1-000958 


1000-347 


0-341 


0-0001482 


1/256 


1000-6527 


1-000476 


1000-176 


0-171 


0-0000740 


1/512 


1000-3264 


1-000237 


1000-089 


0-087 


0-0000379 


1/16 


1010- li 1,38 


1-007568 


1002-864 










Table No. 20. 










RUBIDIUM BROMATE. RbBr03 = 


213-5. 








T = 19-5° and S5-0°C. 






1/16 


1013-3438 


1-010255 


1003-057 




(3-0013258) 


1/32 


1006-6719 


1-005123 


1001-541 


1-516 


0-0006571 


1/64 


1003-3359 


1-002566 


1000-767 


0-774 


0-0003356 


1/128 


1001-6680 


1-001260 


1000-407 


0-360 


0-0001563 


1/256 


1000-8340 


1-000642 


1000-191 


0-216 


00000935 


1/512 


1000-4170 


1-000320 


1000-096 


0-095 


0-0000411 


1/16 


lOlS-S^SS 


1-010162 


1003-149 







1/16 


1016-3125 


1-012756 


1003-511 




(3-0015225^ 


1/32 


1008-1562 


1-006377 


1001-767 


1-744 


0-0007554 


1/64 


1004-0781 


1-003211 


1000-864 


0-903 


0-U003918 


1/128 


1002-0390 


1001617 


1000-421 


0-443 


0-0001920 


1/256 


1001-0195 


1-000784 


1000-235 


0-186 


0000810 


1/512 


1000-5097 


1000375 


1000-134 


0-101 


00000437 


1/16 


1016-3126 


1-012769 


1003-508 







68 



MR J. Y. BUCHANAN ON THE 



A. General Tables, giving the Facts of Observation in the columns 

under m, W, and S. 



TRIAD OF lODATES. 

Table No. 22. 



POTASSIUM lODATE. KI03 = 214-1. 







T = 19-5 


° and 2S-0° C. 








Weight of 
Solution. 


Specific 


Displacement. 


Differences of 


Differences of 

TiOO'nnf'riTns or 




Grams. 


Gi'rtvity. 


G,. 


Displacements. 


Displacements. 


7)1. 


W. 


S. 


W/S = A. 


rfA. 


d log A. 


1/4 


1053-5250 


1-044302 


1008-832 




(3-0038187) 


1/8 


1026-7625 


1-022327 


1004-339 


4-493 


0-0019383 


1/16 


1013-3812 


1-011169 


1002-189 


2-150 


0-0009307 


1/32 


1006-6906 


1-005589 


1001-096 


1-093 


0-0004739 


1/64 


1003-3453 


1-002760 


1000-584 


0-512 


0-0002221 


1/128 


1001-6726 


1-001403 


1000-269 


0-315 


0-0001365 


1/256 


1000-8363 


1-000709 


1000-127 


0-142 


0-0000617 


1/512 


1000-4181 


1-000361 


1000-057 


0-070 


0-0000304 


1/16 


lois-ssm 


1-011U7 


1002-210 










Table No. 23. 










EUBIDIUM lODATE. RbI03 = 260-5. 








T = 19-5° and ^5-()° C. 






1/16 


1016-2812 


1-013677 


1002-576 




(3-0011173) 


1/32 


1008-1406 


1-006856 


1001-276 


1-300 


0-0005636 


1/64 


1004-0703 


1-003405 


1000-661 


0-615 


0-0002669 


1/128 


1002-0351 


1-001690 


1000-344 


0-317 


0-0001377 


1/256 


1001-0175 


1-000827 


1000-190 


0-154 


0-0000669 


1/512 


1000-5087 


1-000436 


1000-072 


0-118 


0-0000510 


1/16 


1016-2812 


1-013625 


1002-618 










Table No. 24. 










CAESIUM IODATE. CsIO3 = 308 


•0. 








T = 19-5°and^5-0° C. 






1/16 


1019-2500 


1-016299 


1002-903 




(3-0012590) 


1/32 


1009-6250 


1-008142 


1001-471 


1-432 


0-0006383 


1/64 


1004-8125 


1-004023 


1000-786 


0-685 


0-0002970 


. 1/128 


1002-4062 


1-001948 


1000-457 


0-329 


0-0001426 


1/256 


1001-2031 


1-000930 


1000-272 


0-185 


0-0000802 


1/512 


1000-6015 


1-000449 


1000-152 


0-120 


0-0000524 


1/16 


1019-2500 


1-016226 


1002-976 







SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 69 

A. General Tables, giving the Facts of Observation in the columns 

under m, W, and S. 



Table No 25. 
POTASSIUM CHLORIDE. 
T = 15-0° C. 



KCl = 74-6. 





Weight of 
Solution. 

Grams. 


Specific Gravity. 


Displacement. 
Gx. 


DifTerences of 
Displacements. 


Differences of 
Logarithms of 
Displacements. 


lit. 

1/2 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 


W. 
1037-3000 
1018-6500 
1009-3250 
10046625 
1002-3312 
LOOl-1656 
1000-5828 


S. 
1-023167 
1-011900 
1-005912 
1-002972 
1-001487 
1-000716 
1-000365 


W/S = 
1013 
1006 
1003 
1001 
1000 
1000 
1000 


= A. 

813 
671 
393 
685 
842 
449 
218 


dA. 

7-142 
3-278 
1-708 
0-843 
0-393 
0231 


d log A. 
(3-0059576) 
0-0030702 
0-0014163 
0-0007399 
0-0003653 
0-0001708 
0-0001003 


Table No. 26. 
SODIUM CHLORIDE. NaCl = 58-5. 
T=15-0° C. 


1/2 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 


1029-2500 
1014-6250 
1007-3125 
1003-6562 
1001-8281 
lOCO-9140 
1000-4570 


1020564 
1-010433 
1-005258 
1-002650 
1-001322 
1-000655 
1000322 


1008-510 
1004-148 
1002-043 
1001-003 
1000-505 
1000-259 
1000-135 


4-362 
2-105 
1-040 
0-498 
0-246 
0-124 


(3-0036805) 
0-0018825 
0-0009114 
0-0004508 
0-0002162 
0-0001069 
00000538 

1 


Table No. 27. 
RUBIDIUM BROMIDE. RbBr = 165-5. 
T = 230° C. 


1/8 

1/16 

1/32 

1/64 

1/128 


1020-6875 
1010-3437 
1005-1718 
1002-5859 
1001-2929 


1-015740 
1-007895 
1-003993 
1-001968 
1-000986 


1004-870 
1002-429 
1001-174 
1000-616 
1000-306 


2-441 
1-255 
0-558 
0-310 


(3-0021101) 
0-0010563 
0-0005441 
0-0002420 
0-0001343 


Table No. 28. 

C^SIUxM BROMIDE. CsBr = 213-0. 

T = 23-0° C. 


1/8 

1/16 

1/32 

1/64 

1/128 


1026-6250 
1013-3125 
1006-6562 
1003-.3281 
1001-6640 


1-020672 
1010386 
1-005246 
1-002634 
1-001332 


1005-832 
1002-896 
1001-403 
1000-692 
1000-331 


2-936 
1-493 
0-711 
0-361 


(3-0025257) 
0-0012698 
0-0006470 
0-0003084 
0-0001563 



70 



MR J. Y. BUCHANAN ON THE 



A. General Tables, giving the Facts of Observation in the columns 

under m, W, and S. 



Table No. 29. 



LITHIUM NITRATE. LiNO3 = 69 0. 
T=19-5° C. 





Weight of 
Solution. 


Specific 


Displacement. 


Differences of 


Differeuces of 
Logarithms of 
Displacements. 




Grams. 


Gravity. 


G,. 


Displacements. 


m. 


W. 


S. 


W/S=-A. 


dL. 


d log A. 


1/2 


1034-5000 


1019707 


1014-507 




(3-0062551) 


1/4 


1017-2500 


1-010033 


1007-145 


7-362 


00031631 


1/8 


1008-6250 


1-005031 


1003-565 


3-580 


0-0015418 


1/16 


1004-3125 


1-002548 


1001-759 


1-806 


0-0007865 


1/32 


1002-1562 


1-001290 


1000-864 


0-895 


0-0003882 


1/64 


1001-0781 


1-000654 


1000-423 


0-441 


0-0001914 


1/128 


1000-5390 


1-000336 


1000-203 


0-220 


0-0000957 






Table No. 30. 










SODIUM NITRATE. NaNO3 = 85-0. 








T = 15-0° and 19-5° C. 






1/16 


1005-3125 


1-003453 


1001-852 




(3-0008039) 


1/32 


1002-6562 


1-001715 


1000-939 


0-913 


0-0003962 


1/64 


1001-3281 


1-000863 


1000-464 


0-475 


0-0002061 


1/128 


1000-6640 


1-000431 


1000-232 


0-232 


0-0001005 


112 


10J^2-5000 


1-027810 


1014.-292 




(3-0061632) 


IjJ, 


1021-2600 


l-OlJlp.23 


1007-027 


7-265 


0-0031217 


118 


1010-6250 


1-007119 


1003-Jf.80 


3-547 


0015325 


1116 


1005-3126 


1-003588 


1001-718 


1-762 


0-0007635 


1132 


1002-6562 


1-001802 


1000852 


0-866 


0-0003752 


im 


1001-3281 


1-000900 


1000-1^7 


0-425 


0-0001847 






Table No. 31. 










POTASSIUM NITRATE. KN03 = 1( 


)1-1. 








T=15-0° C. 






1/2 


1050-5500 


1-030874 


1019-087 




(3-0082113) 


1/4 


1025-2750 


1-015700 


1009-427 


9-660 


0-0041364 


1/8 


1012-6375 


1007974 


1004-623 


4-804 


0-0020717 


1/16 


1006-3187 


1-003966 


1002-343 


2-280 


0-0009865 


1/32 


1003-1593 


1-001974 


1001-183 


1-160 


00005032 


1/64 


1001-5796 


1-000985 


1000-594 


0-589 


00002554 


1/128 


1000-7898 

1 


1-000490 


1000-299 


0-295 


0-0001277 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 71 

A. General Tables, giving the Facts of Observation in the columns 

under m, W, and S. 

Table No. 32. 

RUBIDIUM NITRATE. RbN03 = 147-5. 
T = 23-0°C. 





Weight of 
Solution. 

Grams. 


Specific 
Gravity. 


Displacement. 
Gx. 


Differences of 
Displacements. 


Differences of 
Lfigaiithms of 
Displacements. 


m. 

1/4 
1/8 


W. 
1036-8750 
1018-4375 


S. 
1-025590 
1-013065 


W/S = A. 

1011-003 
1005-303 


5-700 


d log A. 

(3-0047526) 

0-0024556 


1/16 


1009-2188 


1-006584 


1002-617 


2-686 


0-0011618 


1/32 


1004-6094 


1-003354 


1001-251 


1-366 


0-0005922 


1/64 


1002-3047 


1-001731 


1000-573 


0-678 


0-0002941 


1/128 


1001-1523 


1-000955 


1000-197 


0-376 


0-0001631 


1/256 


1000-5762 


1-000404 


1000-172 


0-025 


0-0000109 






Table No. 33. 










CESIUM NITRATE. CsN03 = 195-0. 








T = 23-0°C. 






1/8 
1/16 


1024-3750 
1012-1875 


1017943 
1-009035 


1006-318 
1003-124 


3-194 


(3-0027353) 
0-0013804 


1/32 


1006-0937 


1-004536 


1001-550 


1-574 


0-0006818 


1/64 


1003-0468 


1-002288 


1000-757 


0-793 


0-0003445 


1/128 


1001-5234 


1-001186 


1000-337 


0-420 


0-0001822 


1/256 


1000-7617 


1-000580 


1000-181 


0-156 


0-0000673 






Table No. 34. 








c 


5TR0NTIUM NITRATE. SiiNO^\ = 


= 211-6. 








T=15-0°C. 






1/32 
1/64 


1006-6125 
1003-3062 


1-005344 
1-002673 


1001-261 
1000-631 


0-630 


(3-0005477) 
0-0002734 


1/128 


1001-6531 


1-001351 


1000-302 


0-329 


0-0001430 


1/256 


1000-8265 


1-000666 


1000-160 


0-142 


0-0000615 


1/512 


1000-4132 


1-000329 


1000-084 


0-076 


0-0000331 



72 



MR J. Y. BUCHANAN ON THE 



A. General Tables, giving the Facts of Observation in the columns 

under w, W, and S. 



Table No. 35. 

BARIUM NITRATE. Ba(N03)2 = 261-0. 
T = 150°C. 





Weight of 
Solution. 

Grams. 


Specific 
Gravity. 


Displacement. 

Gt. 


Differences of 
Displacements. 


Differences of 
Logarithms of 
Displacements. 


VI. 

1/32 
1/64 

1/128 
1/256 
1/512 
1/1024 


W. 
1008-1562 
1004-0781 
1002-0390 
1001-0195 
1000-5097 
1000-2548 


S. 
1-006719 
1-003377 
1-001693 
1-000836 
1 000422 
1-000218 


W/S = A. 

1001-427 
1000-699 
1000-345 
1000-183 
1000-088 
1000-036 


dA. 

0-728 
0-354 
0-152 
0-095 
0-052 


d log A. 
(3-0006197) 
0-0003164 
0-0001535 
0-0000700 
0-0000416 
00000224 


Table No. 36. 

BARIUM NITRATE. Ba(N03)2 = 261-0. 
T=19-5'C. 


1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/1024 


1016-3125 

1008-1562 
1004-0781 
1002-0390 
1001-0195 
1000-5097 
1000-2548 


1-013302 

1-006697 
1-003367 
1-001710 
1-000856 
1-000433 
1-000205 


1002-971 
1001-449 
1000-708 
1000-328 
1000-163 
1000-076 
1000-049 


1-522 
0-741 
0-380 
0-165 
0087 
0-027 


(3-0012884) 
0-0006592 
0-0003216 
0-0001651 
0-0000716 
0-0000376 
0-0000115 


Table No. 37. 

LEAD NITRATE. Pb(N03)2 = 331-0. 
T=19-5°C. 


1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/1024 


1020-6875 
1010-3437 
1005-1718 
1002-5859 
1001-2929 
1000-6464 
1000-3232 


1-017788 
1-008947 
1-004504 
1-002250 
1-001128 
1-000577 
1-000300 


1002-849 
1001-384 
1000-664 
1000-335 
1000-165 
1000-069 
1000-023 


1-465 

0-720 
0-329 
0-170 
0-096 
0-046 


(3-0012356) 
0-0006347 
0-0003124 
0-0001429 
0-0000739 
0-0000413 
0-0000203 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS, 73 



§ 27. B. Tables giving particulars relating to the exactness of the determinations of 

the specific gravity given in Tables A, in cases where two hydrometers have 

been used. 

In these tables we have, under S^v ^^^ mean specific gravity derived from s^^ series .of observations made 
with Hydrometer No. 21 ; under Sj-, the mean specific gravity derived from s^^ series of observations with 



Hydrometer No. 17; under Sg, the mean specific gravity derived from Sg series of observations with Hydro- 
meter No. 3 ; and_under S the mean of the sum, s, of these series of observations. Under r^ we have the 
probable error of S calculated by the method of least squares; and under d, the maximum departure of the 
mean of any individual series from S. Numbers under r^ and d represent units in the sixth decimal place. 

Table No. 38. 

POTASSIUM CHLORIDE. KCl = 74-6. 

T = 19-5° and^5-^°C. 



m. 


821- 


^21' 


Si,- 


Sl7- 


§. 


s. 


n- 


d. 


1/2 


1-022986 


4 


1-022969 


4 


1-022977 


8 


2-9 


18 


1/4 


1-011674 


4 


1-011665 


4 


1-011670 


8 


3-0 


20 


1/8 


1-005895 


4 


1-005883 


4 


1-005889 


8 


2-1 


16 


1/16 


1-002980 


3 


1-002967 


3 


1-002973 


6 


3-6 


18 


1/32 


1-001494 


3 


1-001485 


4 


1-001489 


7 


2-2 


14 


1/64 


1-000756 


4 


1-000730 


4 


1-000741 


8 


3-1 


27 


1/128 


1-000368 


3 


1-000361 


3 


1-000365 


6 


1-5 


10 


1/256 


1-000199 


4 


1-000188 


4 


1-000193 


8 


2-2 


14 


1/512 


1-000088 


4 


1-000076 


4 


1-000082 


8 


2-2 


14 


1/16 






1-0029U 


3 






0-3 


1 








Table No. 39. 














RUB 


IDIUM CHLORIDE. ; 


RbCl = 121-0. 














T = 19-5° and ^5-6)° 


C. 








1/2 


1-043150 


4 


1-043138 


4 


1-043144 


8 


2-2 


19 


1/4 


1-021875 


4 


1-021858 


3 


1-021868 


7 


2-7 


13 


1/8 


1-011027 


4 


1-011019 


4 


1-011023 


8 


1-5 


10 


1/16 


1-005530 


4 


1-005531 


4 


1005531 


8 


2-5 


15 


1/32 


1-002776 


4 


1-002767 


3 


1-002772 


7 


2-4 


13 


1/64 


1-001398 


4 


1-001402 


3 


1-001400 


7 


2-5 


13 


1/128 


1-000710 


4 


1-000705 


4 


1-000707 


8 


1-8 


12 


1/256 


1-000349 


3 


1-000350 


4 


1-000350 


7 


1-6 


12 


1/512 


1-000163 


3 


1-000163 


3 


1-000163 


6 


0-6 


3 


1/16 






1-005485 


3 






1-8 


5 








Table No. 40. 














C^ 


SIUM CHLORIDE. G 
T = 19-5° and 23-0° 


sGl= 168-5, 
C. 








1/2 


1-062581 


3 


1-062566 


4 


1-062572 


7 


3-1 


19 


1/4 


1-031742 


4 


1-031734 


3 


1-031739 


7 


1-6 


7 


1/8 


1-015996 


4 


1-015991 


3 


1-015994 


7 


1-5 


12 


1/16 


1-008044 


3 


1-008030 


4 


1-008036 


7 


3-2 


20 


1/32 


1-004040 


4 


1-004029 


3 


1-004035 


7 


1-6 


7 


1/64 


1-002032 


4 


1-002021 


4 


1-002027 


8 


2-0 


14 


1/128 


1-001026 


4 


1-001024 


3 


1-001025 


7 


1-8 


15 


1/256 


1-000517 


4 


1-000511 


4 


1-000514 


8 


1-6 


13 


1/512 


1-000253 


4 


1-000244 


3 


1-000249 


7 


1-7 


12 


1/16 






1-007954 


S 






2-0 


6 



TRANS. ROY. SOC. EDIN., VOL. XLIX. PART I. (NO. 1). 



10 



74 



MR J. Y. BUCHANAN ON THE 



B. Tables giving the Probable Error {vq) of the Mean Specific Gravity (S) ; and the 
greatest Departure (d) of the mean of any individual series from S ; both r^ 
and d being expressed in units of the sixth decimal place. 



Table No. 41. 

POTASSIUM BROMIDE. KBr = 119-1. 
T = 19-5° a.-nd 23-0° C. 



m. 


Sgi- 


S21. 


S17. 


hv 


s. 


s. 


»-o- 


d. 


1/2 


1-041284 


4 


1-041272 


4 


1-041278 


8 


2-8 


17 


1/4 


1-020915 


4 


1-020890 


4 


1-020903 


8 


3-6 


24 


1/8 


1-010.034 


4 


1-010521 


4 


1-010528 


8 


2-4 


18 


1/16 


1-005288 


4 


1-005271 


4 


1-005279 


8 


3-8 


26 


1/32 


1-002649 


3 


1002630 


4 


1-002638 


7 


2-7 


13 


1/64 


1-001313 


3 


1-001301 


4 


1-001306 


7 


3-1 


23 


1/128 


1-000650 


3 


1 -000654 


4 


1-000652 


7 


1-6 


11 


1/2.56 


1-000336 


4 


1-000314 


4 


1-0003-J5 


8 


3-1 


20 


1/512 


1-000163 


2 


1-000153 


2 


1-000158 


4 


6-6 


28 


1/16 






1-005306 


■4 






2-0 


9 








Table No. 42.* 














RUBIDIUM BROMIDE. RbBr = 165-5. 














T=19-5°C. 










m. 


S3. 


S3. 


S17. 


Sl7- 


S. 


s. 


»-o- 


d. 


1/2 


1-061235 


3 


1-061259 


3 


1-061247 


6 


48 


26 


1/4 


1-031081 


3 


1-031080 


3 


1-031081 


6 


1-3 


6 


1/8 


1-015674 


3 


1015665 


4 


1-015669 


7 


6-0 


30 


1/16 


1-007865 


3 


1007870 


3 


1-007863 


6 


1-0 


6 


1/32 


1-003941 


3 


1-003949 


3 


1-003945 


6 


2-7 


17 


1/64 


1-001949 


3 


1-001964 


3 


1-001957 


6 


3-1 


16 


1/128 


1-000980 


2 


1-000988 


2 


1-000984 


4 


3-5 


15 


1/256 


1-000478 


2 


1-000446 


4 


1-000457 


6 


5-1 


22 


1/512 


1-000227 


2 


1-000238 


3 


1-000233 


5 


2-3 


11 


1/1024 


1-000090 


3 


1-000070 


4 


1-000079 


7 


3-9 


28 








Table No. 43. 














QA 


5SIUM BROMIDE. Cs 


>Br= 213-0. 














T = 19-5°C. 










1/2 


1-080943 


4 


1-080923 


3 


1-080935 


7 


3-5 


24 


1/4 


1-041024 


3 


1-040997 


3 


1-041011 


6 


5-5 


17 


1/8 


1-020708 


3 


1-020695 


3 


1-020702 


6 


2-9 


14 


1/16 


1-010407 


3 


1-010411 


3 


1-010409 


6 


2-0 


10 


1/32 


1-005203 


3 


1-005160 


3 


1-005182 


6 


6-8 


34 


1/64 


1-002638 


3 


1-002622 


2 


1-('02631 


5 


3-3 


16 


1/128 


1-001269 


2 


1-001271 


4 


1-001270 


6 


3-9 


18 


1/256 


1 -000595 


3 


1-000619 


! 3 


1-000607 


6 


3-7 


28 


1/512 


1-000317 


3 


1-000296 


3 


1-000308 


6 


5-9 


26 


1/1024 


1 000 135 


3 


1-000155 


I ^ 


1-000145 


6 


3-3 


16 



In Tables Nos. 42, 43, 45, and 46 Hydrometer No. 3 has been used in place of No. 21, 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 



75 



B. Tables giving the Probable Error (ro) of the Mean Specific Gravity (S) ; and the 
greatest Departure (d) of the mean of any individual series from S ; both r^ 
and d being expressed in units of the sixth decimal place. 



34 

12 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/1024 



Table No. 44. 

POTASSIUM IODIDE. KI= 166-1. 
T = 19-5°C. 



821- 


%• 


Sn- 


«17- 


S. 


s. 


1-114619 


1 


1-114617 


2 


1-114617 


3 


1-087124 


3 


1-087124 


3 


1-087124 


6 


1-058929 


2 


1-058929 


3 


1-058929 


5 


1-029912 


1 


1-029904 


2 


1-029906 


3 


1-015104 


2 


1-015103 


3 


1-015104 


5 


1-007593 


2 


1-007583 


2 


1-007588 


4 


1-003794 


2 


1-003786 


2 


1-003790 


4 


1-001906 


2 


1-001893 


• 2 


1-001899 


4 


1-000951 


2 


1-000949 


2 


1-000950 


4 


1-000481 


2 


1-000479 


2 


1-000480 


4 


1-000237 


3 


1-000232 


3 


1-000235 


6 


1-000132 


2 


1-000115 


3 


1-000122 


5 



3-0 

2-8 
2-2 
1-3 
5-3 
2-7 
2-2 
2-8 
2-6 
3-2 
2-0 
3-5 



d. 



17 
10 

6 
22 
12 

9 
10 
10 
13 
13 
14 



Table No. 45. 

RUBIDIUM IODIDE. Ebl = 212-5. 
T-19-5° C. 



1/2 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/1024 



S3. 


S3. 


Si7- 


s„. 


g. 


s. 


n- 


1-078425 


5 


1-078416 


5 


1-078421 


10 


2-6 


1-039781 


3 


1-039774 


3 


1-039778 


6 


4-6 


1-020020 


4 


1-020009 


4 


1-020010 


8 


4-2 


1-010058 


3 


1-010033 


3 


1-010046 


6 


4-2 


1-005038 


3 


1-005032 


3 


1-005030 


6 


5-1 


1-002503 


2 


1-002506 


4 


1-002505 


6 


4-1 


1-001231 


3 


1-001240 


4 


1001237 


7 


4-6 


1-000602 


3 


.1000621 


3 


1-000612 


6 


4-3 


1-000272 


3 


1-000269 


3 


1-000272 


6 


2-5 


1000149 


3 


1-000139 


3 


1-000146 


6 


3-4 



d. 



21 
25 
27 
23 
25 
23 
23 
23 
12 
20 



Table No. 46. 

CESIUM IODIDE. Csl = 260-0. 
T=19-5°C. 



12 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/1024 



1-097435 


3 


1-097420 


3 


1-097427 


6 


2-6 


1-049485 


3 


1-049477 


3 


1-049480 


6 


4-0 


1-024973 


3 


1-024973 


4 


1-024973 


7 


2-1 


1-012527 


3 


1-012532 


3 


1-012529 


6 


2-2 


1-06272 


2 


1-006307 


3 


1-006299 


5 


8-0 


1-003109 


3 


1-003130 


3 


1-003120 


6 


3-7 


1-001549 


3 


1-001543 


3 


1-001546 


6 


1-7 


1-000733 


2 


1-000742 


3 


1-000738 


5 


2-8 


1000281 


3 


1-000263 


3 


1-0C0272 


6 


3-1 


1-000114 


2 


1-000093 


4 


1-000100 


6 


3-9 



15 
19 
13 
11 
35 
21 
11 
13 
16 
15 



7& 



MR J. Y. BUCHANAN ON THE 



B. Tables giving the Probable Error (rp) of the Mean Specific Gravity (S) ; and the 
greatest Departure (d) of the mean of any individual series from S ; both Tq 
and cl being expressed in units of the sixth decimal place. 



Table No. 47. 

POTASSIUM IODIDE. KI= 166-1. 
T = 23-0°C. 



m. 


Sji. 


Hv 


Sir- 


Sy,. 


S. 


s. 


'•o- 


d. 


1/2 


1-058643 


2 


1-058636 


3 


1-058639 


5 


1-8 


9 


14 


1-029724 


3 


1-029705 


2 


1-029717 


5 


3-4 


14 


1/8 


1-014981 


2 


1-014999 


2 


1-014990 


4 


3-6 


12 


1/16 


1-007543 


3 


1-007536 


2 


1-007544 


5 


4-2 


18 


1/32 


1-003767 


2 


1-003754 


2 


1-003761 


4 


2-5 


8 


1/64 


1-001919 


2 


1-001918 


2 


1-001919 


4 


2-9 


12 


1/128 


1-000958 


2 


1000945 


4 


1-000950 


6 


2-4 


9 


1/256 


1-000498 


2 


1-000495 


2 


1-000497 


4 


3-4 


14 








Table No. 48. 














RUBIDIUM IODIDE. Rbl = 212-5. . 














T = 23-0° C. 










1/8 


1-020084 


4 


1-020055 


2 


1-020075 


6 


4-3 


20 


1/16 


1-010091 


2 


1-010092 


3 


1-010092 


5 


0-8 


3 


1/32 


1-005046 


2 


1-005040 


3 


1-005043 


5 


1-7 


8 


1/64 


1-002565 


3 


1-002545 


3 


1-002555 


6 


3-6 


17 


1/128 


1-001275 


3 


1-001279 


3 


1-001277 


6 


1-5 


8 


1/256 


1-000670 


3 


1-000635 


3 


1-000653 


6 


4-5 


24 








Table No. 49. 


•■ 












C 


^SIUM IODIDE. Cs 


[=260-0. 














T = 23-0and^6-O° 


C. 








1/8 


1-025089 


2 


1-025075 


3 


1-025081 


5 


4-1 


15 


1/16 


1-012644 


7 


1-012628 


6 


1-012637 


13 


2-6 


23 


1/32 


1-006332 


3 


1-006355 


2 


1-006341 


5 


4-2 


14 


1/64 


1-003165 


3 


1-003160 


2 


1-003163 


5 


1-3 


7 


1/128 


1-001600 


2 


1-001593 


2 


1-001596 


4 


1-3 


4 


1/256 


1-000826 


2 


1-000804 


3 


1-000814 


5 


4-4 


13 


1116 


1-012635 


u 


1-012606 


S 


1-012623 


7 


4-1 


25 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 77 

B. Tables giving the Probable Error [r^ of the Mean Specific Gravity (S) ; and the 
greatest Departure [d) of the mean of any individual series from S ; both r^ 
and d being expressed in units of the sixth decimal place, 

Tablk No. 50. 

POTASSIUM CHLORIDE. KCl = 74-6. 
T=15-0° C. 



Table No. 52. 

RUBIDIUM BROMIDE. RbBr = 165-5. 

T = 23-0° C. 



m. 


S21. 


%• 


Sl7- 


S],. 


S. 


s. 


n- 


d. 


1/2 


1-023170 


4 


1-023164 


4 


1-023167 


8 


2-0 


20 


1/4 


1-011899* 


3* 


1-011900 


3 


1-011900 


6 


M 


5 


1/8 


1-005912 


4 


1-005912 


3 


1-005912 


7 


1-7 


14 


1/16 


1-002977 


4 


1-002967 


4 


1-002972 


8 


1-6 


14 


1/32 


1-001493 


4 


1-001480 


4 


1-001487 


8 


2-1 


13 


1/64 


1-000723 


4 


1-000706 


3 


1-000716 


7 


3-0 


21 


1/128 


1-000365 


4 


1-000365 


4 


1-000365 


8 


2-0 


10 








Table No. 51. 














SODIUM CHLORIDE. NaCl = 58-5. 














T = 15-0°C. 










1/2 


1-020576 


4 


1-020552 


4 


1-020564 


8 


3-9 


25 


1/4 


1-010436 


4 


1-010429 


3 


1-010433 


7 


2-5 


14 


1/8 


1-005265 


3 


1-005253 


4 


1-005258 


7 


2-4 


19 


1/16 


1-002654 


4 


1-002645 


4 


1-002650 


8 


1-7 


11 


1/32 


1-001331 


4 


1-001312 


4 


1-001322 


8 


2-6 


17 


1/64 


1-000658 


3 


1-000652 


4 


1-000655 


7 


2-6 


14 


1/128 


1-000324 


4 


1-000320 


4 


1-000322 


8 


0-7 


6 



1/8 

1/16 

1/32 

1/64 

1/128 



1-015751 
1-007902 
1-004002 
1-001977 
1-000972 



1-015729 
1-007888 
1-003983 
1-001962 
1-000999 



1-015740 
1-007895 
1-003993 
1-001968 
1-000986 



4 


4-2 


4 


2-7 


4 


3-7 


5 


2-9 


4 


5-2 



11 

8 
10 

8 
14 



Table No. 53. 

CiESIUM BROMIDE. CsBr = 213-0. 

T = 23-0° C. 



1/8 

1/16 

1/32 

1/64 

1/128 



1-020681 
1-010395 
1-005238 
1-002636 
1-001328 



1-020662 
1-010376 
1-005251 
1-002630 
1-001335 



1-020672 
1-010386 
1-005246 
1-002634 
1-001332 



4 


3-9 


4 


3-7 


5 


3-8 


5 


1-9 


5 


1-6 



14 
10 
16 
11 

8 



These observations were made with Hydrometer No. 3. 



78 



MR J. Y. BUCHANAN ON THE 



B. Tables giving the Probable Error (r'o) of the Mean Specific Gravity (S) ; and the 
greatest Departure (d) of the mean of any individual series from S ; both r„ 
and d being expressed in units of the sixth decimal place. 



Table No. 54, 

LITHIUM NITRATE. 
T=19-5°C. 



LiNO3 = 69 0. 



HI. 


821- 


S21- 


Sl7- 


hi- 


S. 


s. 


^0- 


d. 


1/2 


1-019726 


2 


1-019688 


2 


1-019707 


4 


7-4 


20 


1/4 


1-010026 


2 


1-010039 


2 


1010033 


4 


2-5 


7 


1/8 


1-005025 


2 


1-005036 


3 


1-005031 


5 


1-9 


7 


1/16 


1-002535 


2 


1-002561 


2 


1-002548 


4 


5-1 


13 


1/32 


1-001285 


2 


1-001294 


2 


1-001290 


4 


1-8 


5 


1/64 


l-( '00649 


2 


1-000658 


2 


1-000654 


4 


1-8 


5 


1/128 


1-000339 


3 


1-000332 


2 


1000336 


5 


1-6 


6 






Table No. 55. 












SODIUM NITRATE. NaNO3 = 850. 












T = 15-0° and 19-5° C. 








1/16 


1003455 


4 


1-003450 


4 


1-003453 


8 


2-1 


17 


1/32 


1-0017-24 


4 


1001706 


4 


1-001715 


8 


2-7 


20 


1/64 


1-000866 


4 


1-000859 


3 


1-000863 


7 


2-4 


12 


1/128 


1-000437 


4 


1-000425 


4 


1-000431 


8 


2-1 


19 


112 


1-027 8IJ1. 


3 


1027806 


4 


1-027810 


7 


2-1 


15 


114 


lOlJ^lSO 


4 


1-0U115 


4 


1-01^23 


8 


3-1 


19 


118 


1-007122 


4- 


1-007117 


4 


1-007119 


8 


2-J, 


15 


ijie 


1-003593 


4 


1-003583 


4 


1-003588 


8 


1-9 


16 


1IS2 


1-001805 


4 


1-001798 


4 


1-001802 


8 


1-3 


10 


im 


1-000907 


3 


1-000893 


3 


1-000900 


6 


2-6 


15 






Table No. 56. 












POTASSIUM NITRATE, KNO3 = 101-l. 












T=19-5°C. 








1/2 


1-030565 


2 


1-030562 


2 


1-030564 


4 


3-6 


13 


1/4 


1015533 


2 


1-015533 


2 


1-015533 


4 


3-7 


13 


1/8 


1-007880 


3 


1-007866 


3 


1-007873 


6 


2-5 


17 


1/16 


1-003978 


6 


1-003958 


6 


1-003968 


12 


3-1 


22 


1/32 


1002015 


2 


1-002010 


2 


1002013 


4 


4-7 


15 


1/64 


1-001011 


3 


1000996 


3 


1-001004 


6 


2-5 


11 


1/128 


1-000513 


3 


1-000503 


2 


1000509 


5 


3-9 


20 






Table No. 57. 












STRONTIUM NITRATE. Sr(N03)2 = 211-6. 












T=15-0°C. 








1/32 


1-005349 


! 3 


1-005338 


3 


1-005344 


6 


2-4 


15 


1/64 


1-002675 


i 4 


1-002672 


4 


1-002673 


8 


1-4 


9 


1/128 


1-001351 


4 


1-001350 


4 


1-001351 


8 


1-5 


8 


1/2.56 


1 000675 


4 


1-000657 


4 


1-000666 


8 


2-8 


16 


1/512 


1-000336 


4 


1-000323 


4 


1-000329 


8 


2-0 


12 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 79 



B. Tables giving the Probable Error (ro) of the Mean Specific Gravity (S) ; and the 
greatest Departure (d) of the mean of any individual series from S ; both Vq 
and d being expressed in units of the sixth decimal place. 



Table No. 58. 

BAEIUM NITEATE. Ba(N03)2 = 261-0. 

T=15 0°C. 



m. 


S21. 


*21- 


Sn- 


%• 


S. 


s. 


'•o- 


d. 


1/32 


1-006718 


6 


1006719 


6 


1-006719 


12 


1-0 


11 


1/64 


1-003381 


4 


1-003373 


4 


1-003377 


8 


1-8 


10 


1/128 


1001694 


4 


1-001692 


4 


1001693 


8 


1-8 


14 


1/256 


1 -000844 


4 


1-000827 


4 


1-000836 


8 


3-1 


17 


1/512 


1-000424 


4 


1000419 


4 


1000422 


8 


2-1 


14 


1/1024 


1-000221 


3 


1-00U217 


4 


1-000218 


7 


1-8 


13 






Table No. 59. 












BAEIUM NITEATE. Ba(N03)2 = 261-0. 












T = 19-5°C. 








1/16 


1-013304 


6 


1-013303 


7 


1-013302 


13 


1-4 


15 


1/32 


1-006698 


4 


1-006695 


4 


1-006697 


8 


2-1 


17 


1/64 


1-003370 


4 


1 003364 


3 


1-003367 


7 


3-0 


23 


1/128 


1-001717 


4 


1-001703 


3 


1-001710 


7 


2-3 


15 


1/256 


1-000856 


8 


1-000856 


8 


1-000856 


16 


1-3 


17 


1/512 


1-000430 


3 


1-000435 


4 


1-000433 


7 


2-2 


19 


1/1024 


1-000214 


7 


1-000196 


7 


1-000205 


14 


2-4 


23 






Table No. 60. 












LEAD NITRATE. Pb(N03)2 = 331-0. 












T=19-5°C. 








1/16 


1-017791 


3 


1-017784 


3 


1-017788 


6 


2-8 


14 


1/32 


1-0U8948 


3 


1-008946 


3 


1008947 


6 


0-2 


13 


1/64 


1-004508 


4 


1-004502 


3 


1004504 


7 


2-5 


14 


1/128 


1002262 


3 


1 002238 


3 


1-002250 


6 


6-4 


40 


1/256 


1-001134 


3 


1001123 


4 


1-001128 


7 


1-8 


11 


1/512 


1-000589 


4 


1-000564 


4 


1000577 


8 


3-8 


24 


1/1024 


1-000305 


4 


1-000293 


3 


1-000300 


7 


2-1 


13 






Table No. 61. 












POTASSIUM NITEATE. KNO3=101-l. 












T = 15-0°C. 


- 






1/2 


1-030857 


2 


1-030891 


2 


l-030«74 


4 


11-6 


49 


1/4 


1-015701* 


3* 


1-015698 


2 


1-015700 


5 


1-5 


7 


1/8 


1-007971* 


3* 


1-007977 


3 


1007974 


6 


1-7 


12 


1/16 


1-003970 


4 


1003962 


4 


1-003966 


8 


16 


13 


1/32 


1-001980 


4 


1-001968 


4 


1-001974 


8 


1-9 


15 


1/64 


1-000989 


4 


1-000982 


4 


1-000985 


8 


1-6 


12 


1/128 


1000494 


4 


1-000487 


4 


l-000i90 


8 


1-6 


10 



These observation.? were made with Hydrometer No. 3. 



80 



MR J. Y. BUCHANAN ON THE 



28. C. Tables giving a Suminary of the Specific Gravities of the Solutions of 
different Salts at different Temperatures. 

CHLORIDES, BROMIDES, AND IODIDES. 



Table No. 62. 
CHLORIDES. MCI. 



M = 


Na. 


K. 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


T = 


15-0° C. 


19-5° 0. 


23-0° 0. 


m. 


Specific Gravity. 


Specific Gravit; 


r. 


Specific Gravity. 


1/2 


1-020564 


1-023167 


1-022977 


1-043144 


1-062572 








1/4 


1-010433 


1-011900 


1-011670 


1-021868 


1-031739 








1/8 


1-005258 


1-005912 


1-005889 


1-011023 


1-015994 








1/16 


1-002650 


1-002972 


1-002973 


1-005531 


1-008036 


1-002924 


1-005485 


1-007954 


1/32 


1-001322 


1-001487 


1-001489 


1-002772 


1-004035 








1/64 


1-000655 


1-000716 


1-000741 


1-001400 


1 -00-2027 








1/128 


1-000322 


1-000365 


1-000365 


1-000707 


1-001025 








1/256 






1-000193 


1-000350 


1-000514 








1/512 






1-000082 


1-000163 


1-000249 









Table No. 63. 
BROMIDES. MBr. 



M = 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


T= 


19-5° C. 


23 -CC. 


m. 

1/2 

14 

1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/1024 


1-041278 
1-020903 
1-010528 
1-005279 
1-002638 
1-001306 
1-000652 
1-000325 
1-000158 


Specific Gravity. 

1-061247 
1-031081 
1-015669 
1-007868 
1-003945 
1-001957 
1-000984 
1-000457 
1-000233 
1000079 


1-080935 
1-041011 
1-020702 
1-010409 
1-005182 
1-002631 
1-001270 
1-000607 
1-000308 
1-000145 


1-005306 


Specific Gravity. 

1-015740 
1-007895 
1-003993 
1-001968 
1-000986 


1-020672 
1-010386 
1-005246 
1-002634 
1-001332 



Table No. 64. 
IODIDES. MI. 



M = 


K. 


Rb. 


Cs. 


K. 


Kb. 


Cs. 


Cs. 


T = 


19-5° C. 


23-0° C. 


26-0° C. 


m. 

1/2 
14 
1/8 
1/16 
1/32 
1/64 
1/128 
1/256 
1/512 
j 1/1024 


1-058929 
1-029906 
1-015104 
1-007588 
1-003790 
1-001899 
1-000950 
1-000480 
1-000235 
1-000122 


Specific Gravity 

1-078421 
1-039778 
1-020010 
1-010046 
1-005030 
1-002505 
1-0U1237 
1-000612 
1-000272 
1-000146 


1-097427 
1-049480 
1024973 
1-012529 
1-006299 
1-003120 
1-001546 
1-000738 
1-000272 
1-000100 


1-058639 
1-029717 
1014990 
1-007544 
1-003761 
1001919 
1-000950 
1-000497 


Specific Gravity 

1-020075 
1-010092 
1-005043 
1-002555 
1-001277 
1-000653 


1-025081 
1-012637 
1-006341 
1-003163 
1-001596 
1-000814 


Specific Gravity. 
1-012623 



SPECIFIC GRAVITY A^D DISPLACEMENT OF SOME SALINE SOLUTIONS. 81 



C. Tables giving a Summary of the Specific Gravities of the Solutions of 
different Salts at difierent Temperatures. 



Table No. 65. 
NITEATES. M'NOj and M"(N03)2. 



M'orM" = 


Na. 


K. 


Sr". 


Ba". 


Li. 


Na. 


Ba". 


Pb". 


Rb. 


Cs. 


T = 


15-0° C. 


19-5° C. 


23-0 


°C. 


w. 




Specific Gravity. 






Specific Gravit)'. 




Specific Gravity. 


1/2 




1-030874 






1-019707 


1-027810 










1/4 




1-015700 






1-010033 


1-014123 






1-025590 




1/8 




1-007974 






1-005031 


1-007119 






1-013065 


1-017943 


1/16 


1-003453 


1003966 






1-002548 


1-003588 


1-013302 


1-017788 


1-006584 


1-009035 


1/32 


1-001715 


1-001974 


1-005344 


1-006719 


1-001290 


1-001802 


1-006697 


1-008947 


1-003354 


1-0114536 


1/64 


1-000863 


1-000985 


1-002673 


1-003377 


1-000654 


1-000900 


1-003367 


1004504 


1-001731 


1-002288 


1/128 


1-000431 


1-000490 


1001351 


1-001693 


1-000336 




1-001710 


1-002250 


1-000955 


1-001186 


1/256 






1-000666 


1000836 






1-000856 


1-001128 


1-000404 


1-000580 


1/512 






1-000329 


1-000422 






1-000433 


1-000577 






1/1024 








1-000218 






1-000205 


1-000300 







Table No. 66. 
TRIADS OF NITRATES, CHLORATES, BROMATES, AND lODATES. MRO3. 



R0,= 



1/2 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/16 



NO3. 



K. 



Rb. 



Cs. 



19-5° C. 



Specific Gravity. 



1-030564 
1-015533 
1-007873 
1-003968 
1-002013 
1-001004 
1-000509 



-050634 
-025698 
-012973 
-006597 
-003355 
-001750 
-000920 
-000458 



■035619 
017961 
009041 
004585 
002247 
001146 
000604 



CIO,. 



Rb. 



Cs. 



19-5° C. a-ndSS-O'C. 



Specific Gravity. 



-019081 
-009638 
■004863 
-002490 
•001253 
-000633 
-000320 
-000182 



1-004759 



-029153 
-014679 
-007356 
-003691 
-001863 
•000919 
•000459 
■000218 



1-007331 



-039043 
-019686 
-009825 
-004953 
•002409 
-001216 
-000552 
-000210 



1-009886 



BrO,. 



Rb. 



Cs. 



19-5° C. and ^5-0° C. 



Specific Gravity. 



-030144 
-015227 
-007662 
-003846 
•001921 
-000958 
-000476 
-000237 



1-007568 



-010255 
005123 
•002566 
001260 
■000642 
000320 



1-010162 



1-012756 
1-006377 
r003211 
1-001617 
1-000784 
1000375 

1-012759 



10, 



K. 



Rb, Cs 



19-5° C. and ^5-0° C. 



Specific Gravity. 



•044302 
-022327 
-011169 
-005589 
-002760 
■001403 
-000709 
•000361 



1-011147 



-013677 
-006856 
-003405 
-001690 
-000827 
-000436 



1-013625 



1016299 
1-008142 
1-004023 
1001948 
1-000930 
1-000449 

1016226 



TRANS. ROY. SOC. EDIN., VOL. XLIX. PART I. (NO. 1). 



11 



82 



MR J. Y. BUCHANAN ON THE 



C. Tables giving a Summary of the Specific Gravities of the Solutions of 
different Salts at different Temperatures. 

POTASSIUM, RUBIDIUM, AND CESIUM SALTS. 



Table No. 67. 
POTASSIUM SALTS. KR and KRO, 



Ror 
R0,= 


CI. 


NO3. 


CI. 


Br. I. 


NO3. 


CIO3. 


BrOj. 10,. 


Gl. 


1 

Br. j 

1 


1. 


CIO3. 


BrOj. 


10, 


T= 


15-0° C. 


19-5° 0. 


23-0° C. 


m. 


Specific Gravity. 


Specific Gravity. 


Specific Gravity. 


1/2 


1-023167 


1-030874 


1-022977 


1-0412781-058929 


1-030564 












1-058639 






1/4 


1-01 1900 


1-015700 


1-011670 


1-020903,1-029906 


1-015533 


1-019081 


1-030144 


1-044302 






1-029717 








1/8 


1-005912 


1-007974 


1-005889 


1-010528 


1-015104 


1-007873 


1-009638 


1-015227 


1-0223-27 






1-014990 








1/16 


1-002972 


1-0039661-002973 


1-005279 


1-007588 


1-003968 


1-004863 


1 -007662 


1011169 


1-002924 


1-005306 


1-007544 


1-004759 


1-007568 


1-011147 


1/32 


1-001487 


1-001974 


1-0014891-002638 


1-003790 


1-002013 


1-002490 


1-003846 


1-005589 






1003761 








1/64 


1-000716 


1-000985 


1-000741 


1-001306 


1-001899 


1-001 0041 -001 2531 -001921 


1-002760 






1-001919 








1/128 


1-000365 


1-000490 


1-000365 


1-00065211-000950 


1-000509 1-000633 1-000958 


1-001403 






1-000950 








1/256 






1-00U193 


1-0003251-000480 




1-0003201000476 


1-000709 






l-0004<t7 








1/512 






1-000082 


1-000158)1-000235 




1-000182,1-000237 


1-000361 






1 






1/1024 










1-000122 





















Table No. 68. 
RUBIDIUM SALTS. RbR and RbRO,. 



Ror 
R03 = 


CI. 


Br. 


I. 


NO3. 


CIO3. 


BrOj. 


IO3. 


CI. 


Br. 1. 


NO3. 


CIO3. 


Br03. 1 10, 

! ■ 1 


T = 


19-5° 0, 


23-0° C. 1 


m. 


Specific, Gravity. 


Specific Gravity. | 


1/2 


1-043144 


1-061247 


1-078421 


1-050634 




















1 


1/4 


1-021868 


1-031081 


1-039778 


1-025698 


1-029153 












1-025590 






1 


1/8 


1-011023 


1-015669 


1-020010 


1-012973 


1-014679 








1-015740 


1-020075 


1 -013065 








1/16 


1-005531 


1-007868 


1-010046 


1-006597 


1-007356 


1-010255 


1-013677 


1-005485 


1-007895 


1-010092 


1 -006584 


1-007331 


1-010162 


I-OK'Cl'.V 


1/32 


1-002772 


1-003945 


1-005030 


1-003355 


1-003691 


1-005123 


1-006856 




1-003993 


1005043 


1-003354 








1/64 


1-001400 


1-001957 


1-002505 


1-001750 


1-001863 


1-002566 


1-003405 




1-001968 


1-002555 


1-001731 






, 


1/128 


1-000707 


1-000984 


1-001237 


1 -000920 


1-000919 


1 -001260 


1-001690 




1-000986 


1-001277 


1-000955 








1/256 


1-000350 


1-000457 


1-000612 


1-000458 


1-000459 


1-000642 


1-000827 






1-000653 


1 -000404 








1/512 


1-000163 


1-000233 


1-000272 




1-000218 


1-000320 


1-000436 














i 


1/1024 




1-000079 


1-000146 






















1 



Table No. 69 
CESIUM SALTS. CsR and CsEO,. 



Ror 
R03= 



T= 



1/2 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/1024 



CI. 



Br. 


I. 


NO3. 


CIO3. 


Br03. 



10.. 



19-5° C. 



Specific Gravity. 



1-062572 
1-031739 
1-015994 
1-008036 
1 -004035 
1-002027 
1-001025 
1-000514 
1-000249 



•080935 
-041011 
-020702 
•010409 
-005182 
-002631 
-001270 
-000607 
-000308 
•000145 



-097427 
-049480 
-024973 
-012529 
-006299 
-003120 
-001546 
-000738 
•000272 
•000100 



0356191 
0179611 
0090411 
0045851 
002247|1 
0011461 
OOO6O4I1 
1 



039043 
019686| 
0098251 
004953' 1 
00240911 



001216 
000552 
•000210 



-012756 
-006377 
-003211 
-001617 
•000784 
-000375 



■016299 
008142 
■004023 
001948 
000930 
000449 



V/S1\ 


tuu vs tv 


^3- 












CI. 


Br. 


I. 


NO3. 


CIO3. 


BrO.. 


IO3. 


I, 


23-0° C. 


26-0'C, 



Specific Gravity. 



1-007954 



1-020672 1-025081 1-017943 
1-010386 1-012637 1-009035 
1-0052461-006341 1-004536 
1 -002634 1 -003 1 63 1 -002288 
10013321-0015961-001186 
1-0008141-000580 



1009886 



1-012759 



Specific 
Gravity. 



1-016226 1-012625, 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 83 



C. Tables giving a Summary of the Specific Gravities of the Solutions of 
different Salts at different Temperatures. 

Table No. 70. 





The Ennbad, ME: 


-CHLORIDES, BROMIDES, AND IODIDES OF 


POTASSIUM, 








RUBIDIUM, AND CAESIUM. 






M = 




K. 


Rb. 


Cs. 


M= 




K. 


Rb. 


Cs. 1 


T = 




19-5° C. 


T= 




19-5° G. 


m. 


R. 


Specific Gravity. 


m. 


R 


Specific Gravity. 


1/2 


CI 


1-022977 


1-043144 


1-062572 


1/64 


CI 


1-000741 


1-001400 


1-002027 




15r 


1-041278 


1-061247 


1-080935 




Br 


1-001306 


1-001957 


1-002631 




I 


1-058929 


1-078421 


1-097427 




I 


1-001899 


1-002505 


1-003120 


1/4 


CI 


1-011670 


1-021868 


1-031739 


1/128 


CI 


1-000365 


1-000707 


1-001025 




Br 


1-020903 


1-031081 


1-041011 




Br 


1-000652 


1-000984 


1-001270 




I 


1-029906 


1-039778 


1-049480 




I 


1-000950 


1-001237 


1-001546 


1/8 


CI 


1-005889 


1-011023 


1015994 


1/256 


CI 


1-000193 


1-000350 


1-000514 




Br 


1-010528 


1-015669 


1-020702 




Br 


1-000325 


1-000457 


1-000607 




I 


1-015104 


1-020010 


1-024973 




I 


1-000480 


1-000612 


1-000738 


1/16 


CI 


1-002973 


1-005531 


1-008036 


1/512 


CI 


1-000082 


1-000163 


1-000249 




Br 


1-005279 


1-007868 


1-010409 




Br 


1-000158 


1-000233 


1-000308 




I 


1-007588 


1-010046 


1-012529 




I 


1-000235 


1-000272 


1-000272 


1/32 


CI 


1-001489 


1-002772 


1-004035 


1/1024 


CI 










Br 


1-002638 


1003945 


1-005182 




Br 




1-000079 


1-000145 




I 


1-003790 


1-005030 


1-006299 




I 


1-000122 


1-000146 


1-000100 



The Ennead, MRO, 



Table No. 71. 

-CHLORATES, BROMATES, AND lODATES OF POTASSIUM, 
RUBIDIUM, AND CiESIUM. 



M = 




K. 


Rb. 


Cs, 


K. 


Rb. 


Cs. 


T = 




19 5° 0. 


23-0° C. 1 


m. 

1/4 


RO3. 
CIO3 
BrOj 
IO3 




-019081 
-030144 
-044302 


Specific Gravity. 
1-029153 


1-039043 




Specific Gravity. 




1/8 


CIO3 
BrOg 
10., 




-009638 
-015227 
022327 


1-014679 


1-019686 








1/16 


CIO3 
Br03 
IO3 




-004863 
007662 
-011169 


1-007356 
1-010255 
1-013677 


1-009825 
1-012756 
1-016299 


1-004759 
1-007568 
1-011147 


1-007331 
1-010162 
1-013625 


1-009836 
1-012759 
1-016226 


1/32 


CIO3 
Br03 
IO3 




002490 
003846 
005589 


1-003691 
1-005123 
1-006856 


1-004953 
1-006377 
1-008142 








1/64 


CIO3 
BrOg 
IO3 




001253 
001921 
002760 


1-001863 
1-002566 
1-003405 


1-002409 
1-003211 
1-004023 








1/128 


CIO3 
BrOo 
103^ 




000633 
000958 
001403 


1-000919 
1-001260 
1-001690 


1-001216 
1-001617 
1-001948 








1/256 


CIO3 
BrOg 
lOs 




000320 
000476 
000709 


1-000459 
1-000642 
1-000827 


1-000552 
1-000784 
1-000930 








1/512 


CIO3 
BrOj 
IO3 




000182 
000237 
000361 


1-000218 
1-000320 
1-000436 


1000210 
1-000375 
1-000449 









84 



MH J. Y. BUCHANAN ON THE 



29. D. Tables giving a Summary of the Increments of Displacement, v, caused by the 
Dissolution of m grm.-mol. Salt in 1000 grams Water at different Temperatures. 
v = A-\ 000. 

CHLOEIDES, BROMIDES, AND IODIDES. 

Table No. 72. 
CHLORIDES. MCI. 



M = 


Na. 


K. 


K. 


Rb. Cs. 


K. 


Rb. 


Cs. 


T = 


15-0°C. 


19-5° C. 


23-0° 0. 1 


m. 
1/2 
1/4 
1/8 
1/16 
1/32 
1/64 
1/128 
1/256 
1/512 


V. 

8-510 
4-148 
2-043 
1-003 
0-505 
0-259 
0-135 


13-813 
6-671 
3-393 
1-685 
0-842 
0-449 
0-218 


14-001 
6-899 
3-416 
1-684 
0-841 
0-423 
0-217 
0-098 
0-064 


16-637 
8-202 
4-057 
2-020 
1-006 
0-489 
0-238 
0-122 
0'073 


20-401 
10-066 
4-989 
2-475 
1-225 
0-604 
0-291 
0-144 
0-079 


V. 

1-733 


V. 

2-066 


V. 

2-557 



Table No. 73. 
BROMIDES. MBr. 



M = 


K. 


Eb. 


Cs. 


K. 


Rb. 


Cs. 


T-: 


19-5° C. 


23-0° C. 


TO,. 

1/2 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/1024 


17-547 
8-690 
4-314 
2-153 
1-081 
0-554 
0-278 
0-139 
0-074 


20-262 
9-983 
4-941 
2-456 
1-222 
0-627 
0-308 
0-189 
0-090 
0-082 


23-650 
11-756 
5-802 
2-873 
1-466 
0-695 
0-393 
0-224 
0-108 
0-063 


V. 

2-126 


V. 

4-870 
2-429 
1174 
0-616 
0-306 


V. 

5-832 
2-896 
1-403 
0-692 
0-331 



Table No. 74. 
IODIDES. MI. 



M = 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


Cs. 


T = 


19-5° C. 


23-0°C. 


26-0° C. 


VI. 

1/2 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/1024 


22-778 
11-281 
5-574 
2-772 
1-395 
0-695 
0-347 
0-168 
0-089 
0-040 


25-805 
12836 
6-424 
3-203 
1-602 
0813 
0-422 
0-218 
0-143 
0-061 


V. 

29-681 
14-788 
7-343 
3-675 
1-814 
0-939 
0-484 
0-277 
0-235 
0-153 


23-058 
11-467 
5-687 
2-816 
1-424 
0-675 
0-347 
0-152 


V. 

6-360 
3-157 
1-589 
0-763 
0-382 
0-176 


V. 

7-237 
3-608 
1-772 
0-896 
0-434 
0-201 


V. 

3-582 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 85 



D. Tables giving a Summary of the Increments of Displacement, v, caused by the 
Dissolution of m grm.-mol. Salt in 1000 grams Water at different Temperatures. 
t; = A-lOOO. 

Table No. 75. 
NITEATES. M'NOg and M"(N03)2. 



M' or 
M" = 

T = 

m. 


Na. 


K. 


Sr". 


Ba". 


Li. 


Na. 


Ba". 


Pb", 


Rb. 






15- 


0°O. 






19-5 


°0. 




23-0 


°C. 


V. 


V. 


V. 


V. 


V. 


V. 


v. 


V. 


V. 


V. 


1/2 




19'0S7 






14-507 


14-292 










1/4 




9-427 






7-145 


7-027 






11-003 




1/8 




4-623 






3-565 


3-480 






5-303 


6-318 


1/16 


1-852 


2-343 






1-759 


1-718 


2-971 


2-849 


2-617 


3-124 


1/32 


0-939 


1-183 


1-261 


1-427 


0-864 


0-852 


1-449 


1-384 


1-251 


1-550 


1/64 


0-464 


0-594 


0-631 


0-699 


0-423 


0-427 


0-708 


0-664 


0-573 


0-757 


1/128 


0232 


0-299 


0-302 


0-345 


0-203 




0-328 


0-335 


0-197 


0-337 


1/256 






0-160 


0-183 






0163 


0-165 


0-172 


0-181 


1/512 • 


. 




0-084 


0-088 






0-076 


0-069 






1/1024 








0-036 






0-049 


0-023 







Table No. 76. 
TRIADS OF NITRATES, CHLORATES, BROMATES, AND lODATES. MRO, 



R03 = 


NO3. 


CIO3. 


B1O3. 


IO3. 


M = 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


T = 


19-5° C. 


19-5° C. SLudSS-O'C. 


19-5" C. and;?5-0°C. 


19-5° C. and ^5-0° C. 


m. 

1/2 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/16 


19-410 
9-593 
4-727 
2-342 
1-144 
0-575 
0-281 


22-602 
10-897 
5-394 
2-604 
1-250 
0-553 
0-232 
0-118 


I'. 

12-679 
6-301 
3-118 
1-501 
0-798 
0-376 
0-157 


V. 

11-352 
5-632 
2-785 
1-337 
0-661 
0-324 
0-158 
0-057 

2-889 


V. 

12-726 
6-353 
3-183 
1-584 
0-775 
0-400 
0-200 
0-111 

S-204 


V. 

14-515 
7-233 
3-669 
1-804 
0-971 
0-476 
0-292 
0-212 

3-609 


V. 

11-290 
5-576 
2-761 
1-370 
0-688 
0-347 
0-176 
0-089 

2-854 


V. 

3-057 
1-541 
0-767 
0-407 
0-191 
0-096 

3-149 


V. 

3-511 
1-767 
0-864 
0-421 
0-235 
0-134 

3-508 


V. 

8-832 
4-339 
2-189 
1-096 
0-584 
0-269 
0-127 
0-0.^7 

2-210 


V. 

2-576 
1-276 
0-661 
0-344 
0-190 
0-072 

2-618 


V. 

2-903 
1-471 
0-786 
0-457 
0-272 
0-152 

2-976 



86 



MR J. Y. BUCHANAN ON THE 



D. Tables giving a Summary of the Increments of Displacement, i\ caused by the 
Dissolution of m grm.-mol. Salt in 1000 grams Water at different Temperatures. 
t. = A-1000. 

POTASSIUM, RUBIDIUM, AND CAESIUM SALTS. 



Table No. 77. 
POTASSIUM SALTS. Kll and KEO, 



Ror 
R03 = 


01. 


NO3. 


01. 


Br. I. 


NO3. 


CIO3. 1 BrOs. 


IO3. 


01. 


Br. 


I. 


CIO3. 


BrOj. 


10,. 


T = 


15-0° C. 


19-5° 0. 


23-0" 0. 1 


m. 


1'. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


1 

V. 


1/2 


13-813 


19087 


14-001 


17-547 


22-778 


19-410 












23-058 








1/4 


6-671 


9-427 


6-899 


8-690 


11-281 


9-593 


11-352 


11-290 


8-832 






11-467 


" 






1/8 


3-393 


4-623 


3-416 


4-314 


5-574 


4-727 


5-632 


5-576 


4-339 






5-687 








1/16 


1-685 


2-343 


1-684 


2-153 


2-772 


2-342 


2-785 


2-761 


2-189 


1-733 


2126 


2-816 


2-889 


2-854 


2-210 


1/32 


0-842 


1-183 


0-841 


1-081 


1-395 


1-144 


1-337 


1-370 


1-096 






1-424 








1/64 


0-449 


0-594 


0-423 


0-554 


0-695 


0-575 


0-661 


0-688 


0-584 






0-676 








1/128 


0-218 


0-299 


0-217 


0-278 


0-347 


0-281 


0-324 


0-347 


0-269 






0-347 








1/256 






0-098 


0-139 


0-168 




0-158 


0-176 


0-127 






0-152 








1/512 






0-064 


0-074 


0-089 




0-057 


0-089 


0-057 














1/1024 










0-040 























Table No. 78. 
RUBIDIUM SALTS. RbR and RbRO, 



Ror 
R03 = 


01. 


Br. 


I. 


NO3. OIO3. 


Br03. 


IO3. 


01. 


Br. 


I. 


NO3. 


OIO3. 


BrOj. 


10,. 


T= 


19-5° 0. 


23-0° 0. 


m. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


1/2 


16-637 


20-262 


25-805 


22-002 






















1/4 


8-202 


9-983 


12-836 


10-897 


12-726 












11-003 








1/8 


4-057 


4-941 


6-424 


5-394 


6-353 








4-870 


6-360 


5-303 








1/16 


2-020 


2-456 


3-203 


2-604 


3-183 


3-057 


2-576 


2-066 


2-429 


3-157 


2-617 


3-204 


3-149 


2-618 


1/32 


1-006 


1-222 


1-602 


1-250 


1-584 


1-541 


1-276 




1-174 


1-589 


1-251 






1 


1/64 


0-489 


0-627 


0-813 


0-553 


0-775 


0-767 


0-661 




0-616 


0-763 


0-573 








1/128 


0-238 


0-308 


0-422 


0-232 


0400 


0-407 


0-344 




0-306 


0-382 


0-197 








1/256 


0-122 


0-189 


0-218 


0-118 


0-200 


0-191 


0-190 






0-176 


0-172 








1/512 


0073 


0-090 


0143 




0-111 


0-096 


0072 
















1/1024 




0-082 


0-061 

























Table No. 79. 
CESIUM SALTS. CsR and CsRO, 



Ror 
R03= 

T = 
m. 


01. 


Br. 


I. NO3. 


CIO3. 


BtO,. 


IO3. 


01. Br. 


I. 


NO3. 


OIO3. 


Br03. 


IO3. 


Csl. 


19 -5° 0. 


23-0° 0. 


26-0° C. 


V. 


V. 


V. 


V. 


V. 


V. 


V, 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


1/2 


•20-401 


23-650 


29-681 


























1/4 


10-066 


11-756 


14-788 


12-679 


14-515 






















1/8 


4-989 


5-802 


7-343 


6-301 


7-233 








5-832 


7-237 


6-318 










1/16 


2-475 


2-873 


3-675 


3-118 


3-669 


3-511 


2-903 


2-557 


2-896 


3-608 


3-124 


3-609 


3-508 


2-976 


3-582 


1/32 


1-225 


1-466 


1-814 


1-501 


1-804 


1-767 


1-471 




1-403 


1-772 


1-550 










1/64 


0-604 


0-695 


0-939 


0-798 


0-971 


0-864 


0-786 




0-692 


0-896 


0-757 










1/128 


0-291 


0-393 


0-484 


0-376 


0-476 


0-421 


0-457 




0-331 


0-434 


0-337 










1/256 


0-144 


0-224 


0-277 


0-157 


0-292 


0-235 


0-272 






0-201 


0-181 










1/512 


0-079 


0-108 


0-235 




0-212 


0-134 


0-152 


















1/1024 




0-063 


0-153 



























SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 



87 



D. Table giving a Summary of the Increments of Displacement, v, caused by the 
Dissolution of m grm.-mol. Salt in 1000 grams Water at different Temperatures. 
v= A -1000. 

Table No. 80. 

The Ennead, MR :— CHLORIDES, BROMIDES, AND IODIDES OF POTASSIUM, 

RUBIDIUM, AND CESIUM. 



M = 




K. 


Kb. 


Cs. 


K. 


Rb. 


Cs. 


T = 




19-5° C. 


23-0° 0. 


m. 
1/2 


R. 

CI 
Br 
I 


14-001 
17-547 

22-778 


16-637 
20-262 
25-805 


20-401 
23-650 
29-681 


V. 

23-058 


V. 


V. 


1/4 


CI 
Br 

I 


6-899 

8-690 

11-281 


8-202 

9-983 

12-836 


10-066 
11-756 
14-788 


11-467 






1/8 


CI 
Br 
I 


3-416 
4-314 
5-574 


4-057 
4-941 
6-424 


4-989 
5-802 
7-343 


5-687 


4-870 
6-360 


5-832 
7-237 


1/16 


CI 
Br 

I 


1-684 
2-153 
2-772 


2-020 
2-456 
3-203 


2-475 
2-873 
3-675 


1-733 
2-126 
2-816 


2-066 
2-429 
3-157 


2-557 
2-896 
3-608 


1/32 


CI 
Br 
I 


0-841 
1-081 
1-395 


1-006 
1-222 
1-602 


1-225 
1-466 
1-814 


1-424 


1-174 
1-589 


1-403 
1-772 


1/64 


CI 
Br 
I 


0-423 
0-554 
0-695 


0-489 
0-627 
0-813 


0-604 
0-695 
0-939 


0-675 


0-616 
0-763 


0-692 
0-896 


1/128 


CI 
Br 

I 


0-217 
0-278 
0-347 


0-238 
0-308 
0-422 


0-291 
0-393 

0-484 


0-347 


0-306 
0-382 


0-331 
0-434 


1/256 


CI 
Br 
I 


0-098 
0-139 
0-168 


0-122 
0-189 
0-218 


0-144 
0-224 
0-277 


0-152 


0-176 


0201 


1/512 


CI 
Br 
I 


0-064 
0-074 
0-089 


0-073 
0090 
0-143 


0-079 
0-108 
0-235 








1/1024 


CI 
Br 

I 


0-040 


0-082 
0-061 


0-063 
0-153 









88 



MR J. Y. BUCHANAN ON THE 



D, Table giving a Summary of the Increments of Displacement, v, caused by the 
Dissolution of m grm.-mol. Salt in 1000 grams Water at different Temperatures. 
1; = A- 1000. 

Table No. 81. 

The Ennbad, MRO3 :— CHLORATES, BROMATES, AND lODATES OF POTASSIUM, 

RUBIDIUM, AND CiESIUM. 



M = 




K. 


Kb. 


Cs. 


K. 


Rb. 


Cs. 


T = 




19-5° C. 


23-0° C. 


7)1. 

1/4 


RO3. 
CIO3 
BrOo 


V. 

11-352 

11-290 
8-832 


12-726 


14-515 


V. 


V. 


V. 


1/8 


CIO3 
BrOo 


5-632 
5-576 
4-339 


6-353 


7-233 








1/16 


CIO3 
Br03 
IO3 


2-785 
2-761 
2-189 


3-183 
3-057 
2-576 


3-669 
3-511 
2-903 


2-889 
2-854 
2-210 


3-204 
3-149 
2-618 


3-609 
3-508 
2-976 


1/32 


CIO3 
BrO, 
103^ 


1-337 
1-370 
1-096 


1-584 
1-541 
1-276 


1-804 
1-767 
1-471 








1/64 


CIO3 
Br03 • 
IO3 


0-661 

0-688 
0-584 


0-775 

0-767 
0-661 


0-971 
0-864 
0-786 








1/128 


CIO3 
Br03 
IO3 


0-324 
0-347 
0-269 


0-400 
0-407 
0-344 


0-476 
0-421 
0-457 








1/256 


CIO3 
Br03 
IO3 


0-158 
0-176 
0-127 


0-200 
0191 
0-190 


0-292 
0-235 

0-272 








1/512 


CIO3 
BrOg 
IO3 


0-057 
0-089 
0-057 


0-111 
0-096 
0-072 


0-212 
0-134 
0-152 









SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 89 



§ 30, E. Tables giving the Values of v/m, that is, the Mean Increments of Dis- 
placement (Tables D), calculated for the Dissolution of 1 grm.-mol. Salt in 
1000 grams Water at diflferent Temperatures. 

CHLORIDES, BROMIDES, AND IODIDES. 



Table No. 82. 
CHLOEIDES. MCI. 



M = 


Na. 


K. 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


T = 


15-0 


°C. 




19-5° C. 




23-0° C. 


m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/m. 


1/2 


17-02 


27-63 


28-00 


33-27 


40-80 








14 


16-59 


26-68 


27-59 


32-80 


40-26 








1/8 


16-34 


27-14 


27-26 


32-45 


39-91 








1/16 


1605 


26-96 


26-95 


32-32 


39-60 


27-72 


33-06 


40-91 


1/32 


1617 


26-97 


26-92 


32-20 


39-22 








1/64 


16-69 


28-75 


27-12 


31-34 


38-70 








1/128 


17-31 


27-92 


27-86 


30-54 


37-24 








1/256 






25-21 


31-46 


36-94 








1/512 






32-77 


37-47 


40-85 









Table No. 83. 
BROMIDES. MBr. 



Table No. 84. 
IODIDES. ML 



M = 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


T = 


19-5°C. 


23-0° 0. 


m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/m. 


1/2 


35-09 


40-52 


47-30 








1/4 


34-76 


39-93 


47-02 








1/8 


34-51 


39-52 


46-42 




38-96 


46-66 


1/16 


34-45 


39-30 


45-97 


34-01 


38-87 


46-33 


1/32 


34-79 


39-10 


46-93 




37-67 


44-89 


1/64 


35-45 


40-17 


44-49 




39-44 


44-28 


1/128 


35-64 


39 51 


50-38 




39-27 


42-47 


1/256 


35-76 


48-46 


57-54 








1/512 


38-14 


46-18 


55-29 








1/1024 




84-48 


64-51 









M = 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


Cs. 


T = 


19-5° C. 


23-0°C. 


26-0°C. 


m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/jn. 


v/m. 


1/2 


45-55 


51-71 


59-36 


46-11 








1/4 


45-12 


51-34 


5915 


45-86 








1/8 


44-59 


51-39 


58-74 


45-50 


50-88 


57-89 




1/16 


44-25 


51-24 


58-80 


45-06 


50-51 


57-73 


57-32 


1/32 


44-65 


51-28 


58-06 


45-57 


50-86 


56-72 




1/64 


44-58 


52-04 


60-12 


43-20 


48-87 


57-39 




1/128 


44-45 


54-09 


6200 


44-50 


48-97 


55-62 




1/256 


43-18 


55-80 


71-01 


39-91 


45-23 


51-68 




1/512 


45-61 


73-21 


120-67 










1/1024 


41-16 


62-97 


157-49 











TRANS. ROY. SOC. EDIK, VOL. XLIX. PART I. (NO. 1). 



12 



90 



MR J. Y. BUCHANAN ON THE 



E. Tables giving the Values of vjm, that is, the Mean Increments of Displacement 
(Tables D), calculated for the Dissolution of 1 grm.-mol. Salt in 1000 grams 
Water at different Temperatures. 

Table No. 85. 
NITRATES. M'NOa or M"(N03)2. 



M'or 
M"= 


Na. 


K. 


Sr". 


Ba". 


Li. 


Na. 


Ba". 


Pb". 


Rb. 


Cs. 


T = 


15-0° C. 


19-5° C. 


23-0° C. 


m. 
1/2 
1/4 
1/8 
1/Ifi 
1/32 
1/64 
1/128 
1/256 
1/512 
1/1024 


vjm. 

29-64 
30-05 
29-71 
29-79 


vim. 
38-17 
37-71 
36-98 
37-49 
37-85 
38-02 
38-38 


vjm. 

40-37 
40-42 
38-66 
41-08 
43-05 


v/vi. 

45-69 
44-74 
44-16 
47-07 
45-05 
37-37 


v/m. 
29-01 
28-58 
28-52 
28-15 
27-66 
27-10 
25-98 


v/m. 
28-58 
28-11 
27-84 
27-48 
27-28 
27-34 


v/m. 

47-53 
46-39 
45-34 
41-99 
41-72 
39-11 
50-89 


v/m. 

45-58 
44-31 
42-53 
41-59 
42-29 
35-78 
23-55 


vjm. 

44-01 
42-42 
41-88 
40-04 
36-69 
25-26 
43-94 


vjm. 

50-54 
49-98 
49-62 
48-43 
43-13 
46-51 



Table No. 86. 
TRIADS OF NITRATES, CHLORATES, BROMATES, AND lODATES. MRO3. 



R03 = 


NO3. 


CIO3. 


Br03. 


10,. 


M = 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


T = 


19-5° C. 


19-5° 


C. and S3 


-o°c. 


19-5= C. and SS-Q' C. 


19 -5° C. ■and £3-0° C. 


m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/m. 


v/vi. 


v/m. 


1/2 


38-82 


44-00 






















1/4 


38-37 


43-59 


50-71 


45-40 


50-90 


58-06 


45-16 






35-33 






1/8 


37-81 


43-15 


50-41 


45-06 


50-82 


57-86 


44-61 






34-71 






1/16 


37-48 


41-67 


49-89 


44-57 


50-93 


58-71 


44-17 


48-91 


56-69 


35-02 


41-22 


46-45 


1/32 


36-62 


40-05 


48-04 


42-80 


50-69 


57-73 


43-86 


49-31 


56-56 


35-16 


40-85 


47-07 


1/64 


36-80 


35-41 


51-09 


42-34 


49-61 


62-17 


44-07 


49-10 


55-32 


37-37 


42-33 


50-32 


1/128 


35-96 


29-73 


48-21 


41-53 


51-26 


60-95 


44-42 


5209 


54-00 


36-77 


44-07 


58-76 


1/256 




30-25 


40-29 


40-55 


51-37 


74-98 


45-20 


49-04 


59-21 


32-56 


48-69 


69-83 


1/512 








29-63 


57-24 


108-90 


45-67 


49-56 


68-86 


29-23 


37-17 


77-82 


1/16 








46-23 


51-27 


57-75 


45-67 


50-37 


56-13 


35-41 


41-90 


47-61 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 91 

E. Tables giving the Values of vjm, that is, the Mean Increments of Displacement 
(Tables D), calculated for the Dissolution of 1 grm.-mol. Salt in 1000 grams 
Water at different Temperatures. 

POTASSIUM, RUBIDIUM, AND CESIUM SALTS. 

Table No. 87. 













POTASSIUM SALTS. KR and KROj. 












Ror 
R03= 


CI. 


NO3. 


CI. 


Br. 


I. 


NO3. 


CIO3. 


BrO^. 


IO3. 


CI. 


Br. 


I. 


CIO3. 


BrOa. 


IO3. 


T = 


15-0° 0. 


19-5° C. 


23-0° C. 


m. 


vim. 


vjm. 


vjin. 


vjm. 


vhn. 


vjin. 


vjin. 


vjm. 


vim. 


vim. 


v/m. 


vim. 


vim. 


vim. 


vim. 


1/2 


27-63 


38-17 


28-00 


35-09 


45-55 


38-82 












46-11 








1/4 


26-68 


37-71 


27-59 


34-76 


45-12 


38-37 


45-40 


45-16 


35-33 






45-86 








1/8 


27-14 


36-98 


27-26 


34-51 


44-59 


37-81 


45-06 


44-61 


34-71 






45-50 








1/16 


26-96 


37-49 


26-95 


34-45 


44-25 


37-48 


44-57 


44-17 


35-02 


■21-12 


34-01 


45 06 


46-23 


45-67 


35-41 


1/32 


26-97 


37-85 


26-92 


34-79 


44-65 


36-62 


42-80 


43-86 


35-16 






45-57 








1/64 


28-75 


38-02 


27-12 


35-45 


44-58 


36-80 


42-34 


44-U7 


37-37 






43-20 








1/128 


27-92 


38-38 


27-86 


35-64 


44-45 


35-96 


41-53 


44-42 


36-77 






44 50 








1/256 






25-21 


35-76 


43-18 




40-55 


45-20 


32-56 






39-91 








1/512 






32-77 


38-14 


45-61 




29-63 


45-67 


29-23 














1/1024 










41-16 























Table No. 88. 
RUBIDIUM SALTS. RbR and RbRO, 



Ror 
R03 = 


CI. 


Br. 


I. 


NO3. 


CIO3. 


B1O3. 


IO3. 


CI. 


Br. 


I. 


NO3. 


CIO3. 


BrOj. 


IO3. 


T= 


19-5° C. 


23-0° C. 


m. 


vjm. 


vjm. 


vim. 


vim. 


vjm. 


vjm. 


vjm. 


vjm. 


vim. 


vim,. 


vjm. 


vjm. 


vjm. 


vim. 


1/2 


33-27 


40-52 


51-71 


44-00 






















1/4 


32-80 


39-93 


51-34 


43-59 


50-90 












44-01 








1/8 


32-45 


39-52 


51-39 


43-15 


50-82 








38-96 


50-88 


4242 








1/16 


32-32 


39-30 


51-24 


41-67 


5093 


48-91 


41-22 


33-06 


38-87 


50-51 


41-88 


51-27 


50-37 


41-90 


1/32 


32-20 


39-10 


51-28 


4005 


50-69 


49-31 


40-85 




37-67 


50-86 


40-04 








1/64 


31-34 


40-17 


52 04 


35-41 


49-61 


49-10 


42-33 




39-44 


48-87 


36-69 








1/128 


30-54 


39-51 


54-09 


29-73 


51-26 


52-09 


44-07 




39-27 


48-97 


25-26 








1/256 


31-46 


48-46 


55-80 


30-25 


51-37 


49-04 


48-69 






45-23 


43-94 








1/512 


37-47 


46-18 


73-21 




57-24 


49-56 


37-17 
















1/1024 




84-48 


62-97 

























Table No. 89. 
CiESIUM SALTS. CsR and CsROj. 



Ror 
R0,= 



T= 



CI. 



Br. 



NO3. CIO,. BrO 



10,, 



19-5* C. 



■w/m. 
40-80 
40-26 
39-91 
39-60 
39-22 
38-70 
37-24 
36-94 
40-85 



vim. 
47-30 
47-02 
46-42 
45-97 
46-93 
44-49 
50-38 
57-54 
55-29 
64-51 



vjm. 
59-36 
5915 
58-74 
58-80 
58-06 
60-12 
62-00 
71-01 
120-67 
157-49 



vjm. 

50-7] 
50-41 
49-89 
48-04 
51-09 
48-21 
40-29 



vjm. 

5806 
57-86 
58-71 
57-73 
62-17 
60-95 
74-98 
108-90 



vjm. 



56-69 
56-56 
55-32 
54-00 
59-21 
68-86 



vjm. 



46-45 
47-07 
50-32 
58-76 
69-83 
77-82 



CI. 



Br. 


I. 


NO3. 


CIO3. 


BrOj. 



10., 



23-0° C. 



vjm. 



40-91 



vjm,. 



46-66 
46-33 

44-89 
44-28 
42-47 



vjm. 



57- 
57- 
56- 
57- 
55- 
51- 



-89 


50- 


■73 


49- 


-72 


49- 


-39 


48- 


-62 


43- 


-68 


46- 



vjm. 



-54 
-98 
-62 
-43 
•13 
•51 



vjm,. 



57-75 



vjm. 



vjm,. 



56-13 47-61 



26-0° C. 



vjm. 



57-32 



92 



MR J. Y. BUCHANAN ON THE 



E. Table giving the Values of vjm, that is, the Mean Increments of Displacement 
(Tables D), calculated for the Dissolution of 1 grm.-mol. Salt in 1000 grams 
Water at different Temperatures. 

Table No. 90. 

The Ennead, MR :— CHLOEIDES, BROMIDES, AND IODIDES OF POTASSIUM, 

RUBIDIUM, AND CAESIUM. 



M = 




K. 


Rb. 


Cs. 


K. 


Kb. 


Cs. 


T = 




19-5° C. 


23-0° C. 


m. 

1/2 


R. 
CI 
Br 
I 


v/m, 
28-00 
35-09 
45-55 


vim. 
33-27 
40-52 
51-71 


v/m. 
40-80 
47-30 
59-36 


v/m. 
46-11 


v/vi. 


vjm. 


1/4 


CI 
Br 
I 


27-59 
34-76 
45-12 


32-80 
39-93 
51-34 


40-26 
47-02 
59-15 


45-86 






1/8 


CI 
Br 

I 


27-26 
34-51 
44-59 


32-45 
39-52 
51-39 


39-91 
46-42 

58-74 


45-50 


38-96 
50-88 


46-66 
57-89 


1/16 


CI 
Br 

I 


26-95 
34-45 
44-25 


32-32 
39-30 
51-24 


39-60 
45-97 
58-80 


27-72 
34-01 
45-06 


33-06 

38-87 
50-51 


40-91 
46-33 
57-73 


1/32 


CI 
Br 
I 


26-92 
34-79 
44-65 


32-20 
39-10 
51-28 


39-22 
46-93 
58-06 


45-57 


37-67 
50-86 


44-89 
56-72 


1/64 


CI 
Br 

I 


27-12 
35-45 
44-58 


31-34 
40-17 
52-04 


38-70 
44-49 
60-12 


43-20 


39-44 

48-87 


44-28 
57-39 


1/128 


01 
Br 

I 


27-86 
35-64 
44-45 


30-54 
39-51 
54-09 


37-24 
50-38 
62-00 


44-50 


39-27 
48-97 


42-47 
55-62 


1/256 


CI 
Br 

I 


25-21 
35-76 

43-18 


31-46 
48-46 
55-80 


36-94 
57-54 
71-01 


39-91 


45-23 


51-68 


1/512 


CI 
Br 

I 


32-77 
38-14 
45-61 


37-47 
46-18 
73-21 


40-85 

55-29 

120-67 








1/1024 


CI 
Br 

I 


41-16 


84-48 
62-97 


64-51 
157-49 









SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 93 



E. Table giving the Values of vjm, that is, the Mean Increments of Displacement 
(Tables D), calculated for the Dissolution of 1 grm.-mol. Salt in 1000 grams 
Water at different Temperatures. 

Table No. 91. 

The Ennead, MEO3 :— CHLORATES, BROMATES, AND lODATES OF POTASSIUM, 

RUBIDIUM, AND CESIUM. 



M = 




K. 


Rb. 


Cs. 


K. 


Eb. 


Cs. 


T = 




IQ'S'C. 


23-0° C. 


m. 

1/4 


RO3. 
CIO3 
BrOs 
IO3 


vim. 
45-40 
45-16 
35-31 


vjm. 
50-90 


v/m. 
58 06 


v/m. 


v/m. 


v/m. 


1/8 


CIO3 
BrOg 
IO3 


45-06 
44-61 
34-71 


50-82 


57-86 








1/16 


CIO3 
BrO, 
103^ 


44-57 
44-17 
35-02 


50-93 
48-91 
41-22 


58-71 
56-69 
46-45 


46-23 
45-67 
35-41 


51-27 
50-37 
41-90 


57-75 
56-13 
47-61 


1/32 


CIO3 
Br03 
IO3 


42-80 
43-86 
35-16 


50-69 
49-31 
40-85 


57-73 
56-56 
47-07 








1/64 


CIO3 
Br03 
IO3 


42-34 

44-07 
37-37 


49-61 
49-10 
42-33 


62-17 
55-32 
50-32 








1/128 


CIO3 
BrOg 
IO3 


41-53 
44-42 
36-77 


51-26 
52-09 
44-07 


60-95 
54 00 
58-76 








1/256 


CIO3 
BrO, 


40 55 
45-20 
32-56 


51-37 
49-04 
48-69 


74-98 
59-21 
69-83 








1/512 


CIO3 
BiO, 


29-63 
45-67 
29-23 


57-24 
49-56 
37-17 


108-90 
68-86 
77-82 









94 



MR J. Y. BUCHANAN ON THE 



STEONG SOLUTIONS. 

§31. A. General Tables giving, in the columns under m, W, and S, the Facts of 
Observation relating to concentrated Solutions of the Salts of the Ennead 
(K, Rb, Cs, CI, Br, 1), the Specific Gravity of which has been determined with 
the Specific Gravity Bottle or Pyknometer. 

TRIAD OF CHLORIDP]S. 



Table No. 92. 

POTASSIUM CHLORIDE. KCI = 74-6. 

T=19-5°C. 





Weijrlit of 
Solution. 

Grams. 


Specific 
Gravity. 


Displacement. 
G,. 


Differences of 
Displacements. 


Differences of 
Logarithms of 
Displacements. 


m. 
1 
2 
3 
4 


W. 

1074-6 
1149-2 
1223-8 
1298-4 


S. 
1-0449 
10853 
1-1222 
1-1562 


W/S = A. 

1028-42 
1058-87 
1090-52 
1122-93 


rfA. 

30-45 
31-65 
32-41 


d\og A. 

-012672 

-012791 

-012719 

(3-0.50352) 


Table No. 93. 

RUBIDIUM CHLORIDE. RbCl = 121-0 

T=19-5*C. 


1/2 

1 

2 

3 

4 

5 

6 

7 

7-5 


10605 
11210 
1242-0 
1363 
1484-0 
1605-0 
1726-0 
1847-0 
1907 5 


1-0426 
1-0832 
1-1592 
1-2283 
12936 
1-3541 
1-4084 
1-4586 
1-4833 


1017-17 
1034-89 
1071-43 
1109-66 
1147-18 
1185 29 
1225-85 
1266-28 
1285-98 


17-72 
36-54 
38-23 
37-52 
3811 
40-56 
40-43 
19-70 


-007503 
-015066 
-015228 
-014442 
-014190 
•014613 
-014092 
-006704 
(3-109235) 1 


Table No. 94. 

CESIUM CHLORIDE. CsCl = 168-5. 

T=19-5°C. 


1/2 

1 

2 

3 

4 

5 

6 

7 

8 

9 


1084-25 

1168-5 

13370 

1505-5 

1674-0 

1842-5 

2011-0 

2179-5 

2348-0 

2516-5 


] 0616 
1-1188 
1-2270 
1-3245 
1-4113 
1-4956 
1-5670 
1-6374 
1-7025 
1-7590 


1021-37 
1044-39 
1089-61 
1136-59 
1186-06 
1231-88 
1283-31 
1331-03 
1378-81 
1430-59 


23-02 
45-22 
46-98 
49-47 
45-82 
51-43 
47-72 
47-78 
51-78 


-009679 
-018408 
-018333 
-018502 
-016462 
-017763 
-015856 
-015316 
-016010 
(3-155515) 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 95 



STRONG SOLUTIONS. 

A. General Tables giving the Facts of Observation in the columns under 

m, W, and S. 



TRIAD OF BROMIDES. 

Table No. 95. 

POTASSIUM BROMIDE. KBr = 119-1. 

T = 19-5°C. 



m. 
1 
2 
3 
4 
5 



Weight of 
Solution. 

Grams. 



W. 
1119-1 
1238-2 
1357-3 
1476-4 
1595-5 



Specific 
Gravity. 



S. 
1-0808 
1-1545 
1-2220 
1-2843 
1-3425 



Displacement. 

Gt. 



W/S = A. 

1035-44 
1072-49 
1110-71 
1149-57 
1188-16 



Differences of 
Displacements. 



37 05 

38 22 
38-87 
38-59 



Differences of 
Logarithms of 
Displacements. 



d log A. 

-015273 
■015207 
■014932 
-014336 
(3074873) 



Table No. 96. 

RUBIDIUM BROMIDE. RbBr=165-5. 

T = 19-5°C. 



1/2 

1 

2 

3 

4 

5 

6 



1082-75 


1- 


1165-5 




1331-0 




1496-5 




1662-0 




1827-5 


1 • 


1993-0 





•0613 
-1193 
-2281 
-3272 
-4178 
•5009 
-5772 



102021 
1041-26 
1083-71 
1127-56 
1172-23 
1217-56 
1263-57 



21-05 


•008869 


42-45 


-017353 


43-85 


-017226 


44-67 


-016873 


45-33 


-016477 


46-11 


-016108 




(3-101599) 



Table No. 97. 

CESIUM BROMIDE. CsBr = 213-0. 

T = 21-4"' C. 



1213-0 
14260 
1639^0 
18520 
2065^0 



1-1590 
1-3041 
1-4326 
1-5548 
1-6624 



1046-54 
1093-45 
1144-02 
1191-10 
1242-16 



46-91 
50-57 
47-08 
51-06 



-019046 
-018696 
-017495 
-018232 
(3-094181) 



96 



MR J. Y. BUCHANAN ON THE 



STRONG SOLUTIONS. 

A. General Tables giving the Facts of Observation in the columns under 

m, W, and S. 

tkiad of iodides. 

Table No. 98. 

POTASSIUM IODIDE. KI = 166-1. 

T = 19^5° C. 





Weight of 
Solution. 

Grams. 


Specific 
Gravity. 


Displacement. 
G,. 


Diflferences of 
Displacements. 


Differences of 
Logarithms of 
Displacements. 


m. 
1 
2 
3 
4 
5 
6 
7 
8 


W. 
1166-1 
13322 
1498-3 
1664-4 
1830-5 
1996-6 
2162-7 
2328-8 


S. 
1-1146 
1-2177 
1-3128 
1-3982 
1-4766 
1^5483 
1-6141 
1-6749 


W/S = A. 
1046-20 
1094-03 
1141-30 
1190-39 
1239-67 
1289-54 
1339-88 
1390-54 


dA. 

47-83 
47-27 
49-09 
49-28 
49-87 
50^34 
50-53 


d log A, 

•019412 
-018371 
-018288 
-017618 
•017129 
•016630 
•016077 
(3^143143) 


Table No. 99. 

EUBIDIUM IODIDE. Rbl = 212-5. 
T = 19^5° C. 


1/2 

1 

2 

3 

4 

5 

6 

7 


1106-25 

1212-5 

1425-0 

1637-5 

18500 

2062-5 

2275-0 

2487^5 


1-0771 
1-1498 
1-2835 
1-4026 
1-5102 
1-6081 
1-6962 
1-7705 


1026-99 
1054-45 
1110-22 
1167-45 
1224-93 
1282-52 
1341-22 
1404-94 


27-46 
55-77 
57-23 
57-48 
57-59 
58-70 
63-72 


•011459 
•022383 
•021829 
•020873 
•019952 
-019435 
-020157 
(3-147657) 


Table No. 100. 

CESIUM IODIDE. Csl = 260^0. 
T = 23^r C, 


1 
2 
3 

S'S85 


1260^0 
1520-0 
1780-0 

1880-1 


1-1847 
1-3463 

1-4924 

1-5426 


1063-55 
1128 96 
1192-70 

1218-76 


65^41 
63-74 


-025920 

-023852 

(3-076531) 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 97 



STRONG SOLUTIONS. 

A. General Tables giving the Facts of Observation in the columns under 

m, W, and S. 



Table No. 101. 

RUBIDIUM NITRATE. RbN03 = 147-5. 
T = 19-5°C. 





Weight of 

Solution. 

Grams. 


Specific 
Gravity. 


Displacement, 
G,. 


Differences of 
Displacements. 


Differences of 
Logarithms of 
Displacements. 


m. 
1/2 

1 

2 

3 


W. 
1073-75 
1147-5 
1295-0 
1442-5 


S. 
1-0505 
1-0982 
11861 
1-2655 


W/S = A. 
1022-13 
1044-89 
1091-81 
1139-87 


dA. 

22-76 
46-92 
48-06 


d log A. 

-009564 

-019077 

-018705 

(3-056853) 


Table No. 102. 

LITHIUM NITRATE. LiNO3 = 69-0. 
T = 19-5° C. 


1 
2 
3 
4 
5 
6 
7 
8 
9 
10 


1069-0 
11380 
1207-0 
1276-0 
1345-0 
1414-0 
14830 
1552-0 
1621-0 
1690-0 


1-0389 
1-0747 
M081 
1-1392 
1-1684 
M959 

r2457 
1-2684 
1-2906 


1028-97 
1058-90 
1089-25 
1120-08 
1151-14 
1182-37 
1214-18 
1245-88 
1277-98 
1309-47 


29-93 
30-35 
30-83 
30-89 
31-23 
31-81 
31-70 
32-10 
31-49 


-012451 
-012273 
-012122 
-011880 
•011623 
•011528 
•011194 
•011048 
•010568 
(3^117095) 


Table No. 103. 

SODIUM NITRATE. NaNO3 = 850. 
T = 19-5°C. 


1 
2 
3 
4 
5 
6 
7 
8 
9 


10850 
1170-0 
1255-0 
13400 
1425-0 
15100 
15950 
1680-0 
1765-0 


1-0542 
1-1030 
1-1478 
1-1887 
1-2267 
1-2613 
1-2939 
1-3253 
1-3539 


1029-18 
1060-72 
1093-36 
1124-64 
1161-63 
1197-11 
1232-63 
1267-57 
1303-56 


31-54 
32-64 
31-28 
36-99 
35-48 
35-52 
34^94 
35-99 


013109 
•013162 
-012250 
-014054 
-013066 
-012698 
-012139 
-012159 
(3-115131) 



TKANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 



13 



98 



MR J. Y. BUCHANAN ON THE 



STRONG SOLUTIONS. 



§ 32. Tables of the Classes C, D, and E, giving Summaries of their Specific Gravities 
(S), their Increments of Displacement, v, and their Mean Increments of Dis- 
placement per gram-molecule Salt, v/m, respectively. 



R= 


CHLORIDES. 




BROMIDES. 




IODIDES. 


M = 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


T = 


19-5° C. 


19-5° C. 


21 -4° C. 


19-5' C. 


23 -rc. 


Table No. 104. 


Table No. 105. 


Table No. 106. 


C. Specific Gravity. 


C. Specific Gravity. 


C. Specific Gravity. 


m. 


S. 


S. 


S. 


S. 


S. 


S. 


S. 


S. 


s. 


1/2 




1-0426 


1-0616 






1-0613 








1-0771 




1 


1-0449 


1-0832 


1-1188 




1-0808 


1-1193 


1-1590 




1-1146 


1-1498 


1-1847 


2 


1-0853 


1-1592 


1-2270 




1-1545 


1-2281 


1-3041 




1-2177 


1-2835 


1-3463 


3 


1-1222 


1-2283 


1-3245 




1-2220 


1-3272 


1-4326 




1-3128 


1-4026 


1-4924 


4 


1-1562 


1-2936 


1-4113 




1-2843 


1-4178 


1-5548 




1-3982 


1-5102 




5 




1-3541 


1-4956 




1-3425 


1-5009 


1-6624 




1-4766 


1-6081 




6 




1-4084 


1-5670 






1-5772 






1-5483 


1-6962 




7 




1-4586 


1-6374 












1-6141 


1-7705 




8 






1-7025 












1-6749 






9 






1-7590 


















Table No. 107. 


Table No. 108. 


Table No. 109. 


D. Increment of Displacement. 


D. Increment of Displacement. 


D. Increment of Displacement. 


m. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


V. 


1/2 




17-17 


21-37 






20-21 








26-99 




1 


28-42 


34-89 


44-39 




35-44 


41-26 


46-54 




46-20 


54-45 


63-55 


2 


58-87 


71-43 


89-61 




72-49 


83-71 


93-45 




94-03 


110-22 


128-96 


3 


90-52 


109-66 


136-59 




110-71 


127-56 


144-02 




141-30 


167-45 


192-70 


4 


122-93 


147-18 


186-06 




149-57 


172-23 


191-10 




190-39 


224-93 




5 




185-29 


231-88 




188-16 


217-56 


242-16 




239-67 


282-52 




6 




225-85 


283-31 






263-57 






289-54 


341-22 




7 




266-28 


331-03 












339-88 


404-94 




8 






378-81 












390-54 






9 






430-59 
















1 




Table No. 110. 




Table No. 111. 


Table No. 112. 


E. 


Mean Increment of Displac 


ement per 


E. Jlean Increment of Displacement 


E. Mean Increment of Displacement 




gram-molecule. 






per gram-molecule. 




per gram-molecule. 


m. 


vjm. 


v/m. 


v/m. 


vim. 


vjm,. 


v/m. 


v/m. 


v/m. 


v/m. 


1/2 




34-34 


42-74 






40-42 








53-98 




1 


28-42 


34-89 


44-39 




35-44 


41-26 


46-54 




46-20 


54-45 


63-55 


2 


29-43 


35-72 


44-80 




36-25 


41-85 


46-72 




47-02 


55-11 


64-48 


3 


3017 


36-55 


45-53 




36-90 


42-52 


48-00 




47-10 


55-81 


64-23 


4 


30-73 


36-79 


46-51 




37-39 


43-05 


47-77 




47-59 


56-23 




5 




3706 


46-37 




37-63 


43-51 


48-43 




47-93 


56-50 




6 




37-64 


47-21 






43-92 






48-26 


56-87 




7 




38-04 


47-29 












48-55 


57-84 




8 






47-35 












48-81 






9 






47-84 



















SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 99 



STRONG SOLUTIONS. 

Tables of the Classes C, D, and E, giving Summaries of their Specific Gravities (S), 
their Increments of Displacement, v, and their Mean Increments of Displace- 
ment per gram-molecule Salt, v/m, respectively. 



POTASSIUM SALTS. KR. 



R= 
T= 


01. 


Br. 


I. 


19 -5° C. 


Table No. 113. 


C. Specific Gravity. 



S. 


S. 




1-0449 


1-0808 




1-0853 


1-1545 




1-1222 


1-2220 




1-1562 


1-2843 






1-3425 


1 . 

l-( 

K 



s. 

[•1146 
L-2177 
[-3128 
[-3982 
[-4766 
L-5483 
[•6141 



Table No. 116. 



D. Increment of Displacement. 



28-42 

58-87 

90-52 

122-93 



35-44 

72-49 

110-71 

149-57 

188-16 



46-20 
94-03 
141-30 
190-39 
239-67 
289-54 
339-88 
39054 



Table No. 119. 



E. Mean Increment of Displacement 
per gram-molecule. 



m. 


v/m. 


v/m. 


1 


28-42 


35-44 


2 


29-43 


36-25 


3 


30-17 


36-90 


4 


30-73 


37-39 


5 




37-63 


6 






7 






8 







v/m. 

46-20 

47-02 

47-10 

47-59 

47-93 

48-26 

48-55 

48-81 



RUBIDIUM SALTS. RbR. 



R= 

T = 


CI. 


Br. 


I. 


19-5° C. 


Table No. 114. 


C. Specific Gravity. 


VI. 


S. 


S. 


S. 


1/2 


1-0426 


1-0613 


1-0771 


1 


1-0832 


1-1193 


1-1498 


2 


1-1592 


1-2281 


1-2835 


3 


1-2283 


1-3272 


1-4026 


4 


1-2936 


1-4178 


1-5102 


5 


1-3541 


1-5009 


1-6081 


6 


1-4084 


1-5772 


1-6962 


7 


1-4586 




1-7705 


7-5 


1-4833 






Table No. 117. 


D 


. Increment of Displac 


ement. 


m. 


V. 


V. 


V. 


1/2 


17-17 


20-21 


26-99 


1 


34-89 


41-26 


54-45 


2 


71-43 


83-71 


110-22 


3 


109-66 


127-56 


167-45 


4 


147-18 


172-23 


224-93 


5 


185-29 


217-56 


282-52 


6 


225-85 


263-57 


341-22 


7 


266-28 




404-94 


7-5 


285-98 






Table No. 120. 


E. I 


►lean Increment of Disp 
per gram-molecule. 


lacemeut 


■m. 


v/m. 


vjm. 


v/m. 


1/2 


34-34 


40-42 


53-98 


1 


34-89 


41-26 


54-45 


2 


35-72 


41-85 


55-11 


3 


36-55 


42-52 


55-81 


4 


36-79 


43-05 


56-23 


5 


37-06 


43-51 


56-50 


6 


37-64 


43-92 


56-87 


7 


38-04 




57-84 


7-5 


38-13 







CESIUM SALTS. CsR. 



R = 


CI. 


Br. 


I. 


T = 


ig-s-c. 


21-4° 0. 


23-1* C. 


Table No. 115. 


C. Specific Gravity. 


m. 


s. 


S. 


S. 


1/2 


1-0616 






1 


1-1188 


1-1590 


11847 


2 


1-2270 


1-3041 


1-3463 


3 


1-3245 


1-4326 


1-4924 


4 


1-4113 


1-5548 




5 


1-4956 


1-6624 




6 


1-5670 






7 


1-6374 






8 


1-7025 






9 


1-7590 






Table No. 118. 


D 


Increment of Displacement. 


»i. 


V. 


V. 


V. 


1/2 


21-37 






1 


44-39 


46-54 


63-55 


2 


89-61 


93-45 


128-96 


3 


136-59 


144-02 


192-70 


4 


186-06 


191-10 




5 


231-88 


242-16 




6 


283-31 






7 


331 03 






8 


378-81 






9 


430-59 






Table No. 121. 


E. IV 


lean Increment of Disp^ 
per giani -molecule. 


acement 


m. 


v/m. 


v/m. 


v/m. 


1/2 


42-74 






1 


44-39 


46-54 


63-55 


2 


44-80 


46-72 


64-48 


3 


45-53 


48-00 


64-23 


4 


46-51 


47-77 




5 


46-37 


48-43 




6 


47-21 






7 


47-29 






8 


47-35 






9 


47-84 







100 



MR J. Y. BUCHANAN ON THE 



STRONG SOLUTIONS. 

Tables of the Classes C, D, and E, giving Summaries of their Specific Gravities (S), 
their Increments of Displacement, v, and their Mean Increments of Displace- 
ment per gram-molecule Salt, v/m, respectively. 



R= 


NITRATES. 


M = 


Li. 


Na. 


Rb. 


T= 


19 -5° C. 


Table No. 122. 


1 

C. Specific Gravity. 


m. 


S. 


S. 


S. 


1/2 






1-0505 


1 


r0389 


1-0542 


1-0982 


2 


1-0747 


1-1030 


1-1861 


3 


1-1081 


11478 


1-2655 


4 


1-1392 


1-1887 




5 


1-1684 


1-2267 




6 


1-1959 


1-2613 




7 


1-2214 


1-2939 




8 


1-2457 


1-3253 




9 


1-2684 


1-3539 




10 


1-2906 







NITRATES. 



Li. 



Na. 



Rb. 



19-5° C. 



Table No. 123. 



D. Increment of Displacement. 



28-97 
58-90 
89-25 
120-08 
151-14 
182-37 
214-18 
245-88 
277-98 
309-47 



29-18 
60-72 
93-36 
124-64 
161-63 
197-11 
232-63 
267-57 
303-56 



V. 

22-13 

44-89 

91-81 

139-87 



NITRATES. 



Li, 



Na. 



Rb. 



19-5° C. 



Table No. 124. 



E. Mean Increment of Displacement 
per gram-molecule. 



vim. 

28-97 
29-45 
29-75 
3002 
30-23 
30-39 
30-59 
30-73 
30-89 
30-95 



vim. 

29-18 
30-36 
31-12 
31-16 
32-33 
32-85 
33-23 
33-45 
33-73 



vim. 

44-26 

44-89 

45-90 

46-62 



Section VI. — General Description of Tables. 

§ 33. In the tables giving the results of the experiments made with solutions of a 
particular salt, the weights given are those which would have been used if the 
weighings had actually been made in a vacuum ; and the standard temperature, T, 
at which all the operations have been made, is given at the top with the name of the 
salt, both being constants. 

Of the variables, we have under m the quantity of the salt, expressed as the number, 
whole or fractional, of gram-molecules, which is dissolved in 1000 grams of water, under 
W the weight in grams of the solution so produced, and under S the specific gravity 
of the solution referred to that of distilled water as unity, both having the standard 
temperature T. 

With the exception of the determination of the temperature, the result of every 
series of operations depends only on determinations of weight, and they are independent 
of the work of others. Even the tyro has no difficulty in being assured of the true 
weight of the hydrometer when floating up to the same mark in the solution and in 
distilled water respectively. The difficulty which requires manipulative skill, laboratory 
experience, and perseverance to overcome, is to satisfy the condition that the 
temperatures of the solution and of the distilled water respectively, and that of the 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 101 

hydrometer when immersed in them, are identical, and are really the temperature 
shown by the thermometer, and that this temperature is exactly that chosen as the 
standard for the series of experiments. It requires much study and practice in a 
suitable room before even an experienced chemist or physicist can feel confident that 
he can produce this combination of equalities when required. 

It is to the failure to perceive the necessity of this preliminary education that, 
though the method has been the property of science for forty years, it has been used 
practically by none except myself and those whom I have personally instructed. 

§ 34. In the tables of Class A all the facts of observation are to be found. In all the 
solutions the quantity of water is the same, namely, 1000 grams; the quantity of salt 
dissolved in this mass of water is specified for each solution of the same salt in the 
first column under m, in terms of the gram-molecule. In the second column, under W, 
we have the weight in grams of each solution ; it is given by the sum 1000 + m. MR = W ; 
where MR represents the molecular weight of the salt. The symbol used to express the 
weight of salt dissolved in 1000 grams of water is tv, whence iv = mM^. In the third 
column, under 8, we have the specific gravity of the solution. The experimental data on 
which it is founded are the weight, H', in grams, of the hydrometer when it floats at 
a given division of the stem in the solution, and its weight, H, when it floats at the same 
division in distilled water, both of these liquids having the same temperature, T. The 
quotient H'/H is the specific gravity, S, of the solution, as entered in the third column 
of the tables. If we divide the weight of the solution, W, by its specific gravity, 8, we 
obtain the displacement of the solution, which is entered in the fourth column under 
A. It is the expression of the proportion H : H' : : A : W, in which H and A are weights 
of distilled water, and H' and W are weights of the solution. It may be expressed 
in words as follows : — The displacement of the solution. A, bears to its weight, W, the 
same relation as the weight H of distilled water displaced by the hydrometer bears 
to H', that of the solution displaced by the same portion of the same hydrometer 
at the same temperature, T. Therefore, the unit of displacement used in the tables 
is the space occupied by 1 gram of distilled water having the temperature T. 

Generally, our measure of displacement of a body having the temperature T is the 
weight of distilled water having the same temperature which the body displaces 
when totally immersed in the water. The body in question may be solid, liquid, or 
gaseous. The unit of displacement is then the unit of weight, gram or kilogram, of 
water having the particular temperature, T, which is chosen to suit the conditions 
of the experiment, and it must be the common temperature of the body and the 
water. Under this convention the unit of displacement is the space occupied by, say, 
1 gram water at T, whatever value T may have. Thus, in our experiments the value of 
T is in some 15°, in some 19-5°, and in others 23° ; but whichever temperature is used 
as that of the distilled water, it is also that of the salt or saline solution which is supposed 
to displace it when its specific gravity is being determined. The use of displacement 
instead of volume to specify the amount of space occupied by a body is advantageous 



102 MK J. Y. BUCHANAN ON THE 

only when there is a common temperature and it can be accepted as constant. In 
cases where the temperature is subject to variation, the specification of displacement 
must be by volume, because a weight is not affected by change of temperature. 

It is convenient to have a symbol to place after a number in order to indicate 
that its unit is that of displacement as specified above. It is to be used in cases 
corresponding to those in which the symbol c.c. is used when we express volumes in 
cubic centimetres. A suitable symbol for the unit of displacement is Gx or Gj , in 
which G is the unit of weight and T or Ms the common temperature of the body 
and of the water displaced by it. 

In this research the unit of weight used is the gram, so that our unit of displace- 
ment expresses the space occupied by 1 gram of water at the temperature T. When 
the unit of weight used is the kilogram the symbol becomes K^ . In naval architecture 
the displacement of a ship is always expressed in tons, that is, tons of water of ordinary 
atmospheric temperature. In this research the units of displacement used are expressed 
by the symbols Gja-, G19.5., and Gis-. If the adopted value of T were 4° C, then the 
unit of displacement would be G40, and this is the gravimetric symbol for the standard 
cubic centimetre. 

In the fifth column we have the values of d^, the differences of consecutive values of 
A. The entries in this column have a peculiar interest owing to the fact that the values 
of m which indicate the concentration of the solutions form an ascending geometrical 
series with the common ratio 2. The quantity of water, 1000 grams, is the same in all 
the solutions. If we consider any two consecutive values of A, for instance, A^^g and 
\ii, the increment of displacement produced by dissolving 1/8 MR in 1000 grams of 
water is A^^^j— 1000, and the increment produced by dissolving a further quantity of salt 
equal to 1/8 MR in the solution the displacement of which is Aj^^, is c^A = Aj,^ — Aj,8. 
These increments of displacement have been produced by equal quantities of the 
salt, which has been dissolved in the first case in 1000 grams of distilled water, and 
in the second case in (1/8 MR+ 1000) grams of the solution so produced. If the corre- 
sponding values of (A,„— 1000) and [A.^^^ — A^^) be studied, they will be found to be 
almost always different. Considering only the first nine tables relating to the salts 
of the ennead MR, we see that the difference of these increments {A^^ — A,„) — 
(A„^— 1000) is positive for values of m= 1/16 and higher. It changes sign for a value 
of m lying between 1/16 and 1/32 in the case of KBr, KI, and Rbl ; for m lying 
between 1/32 and 1/64 in the case of KCl, RbBr, and Csl ; for m lying between 1/64 
and 1/128 in the case of CsBr ; for m lying between 1/128 and 1/256 in the case of 
RbCl ; and for m lying between 1/256 and 1/512 in the case of CsCl. 

The main object with which this experimental research was begun was to ascertain 
if such a change of sign occurs at any concentration. It was only by using the hydro- 
metric method that the question could be answered. In the sixth column we have 
the values of d log A, the consecutive differences of the log-displacement. The space 
in this column corresponding to the highest value of m is occupied in brackets by the 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 103 

logarithm of the corresponding highest value of A. With this, and the corresponding 
values of c? log A, the log-displacement of each solution can be at once obtained. 

In Class A there are thirty-seven tables ; of these, twenty-four relate to solutions 
of chlorides, bromides, iodides, and nitrates of the alkalies and alkaline earths, the 
specific gravities of which have been determined with two hydrometers. The values 
of S given for each value of m in each of these tables is the mean of two groups 
of series of nine observations each, each group being made with a different hydro- 
meter. The hydrometers chiefly used were Nos. 17 and 21, and for each value of m 
either three or four series of observations were made with each of these hydrometers. 
The mean of each series is the mean of nine independent values of the specific gravity, 
so that the final mean, S, is the mean of 72 independent observations when four 
series have been made with each hydrometer, and the mean of 54 independent 
observations when three series have been made with each hydrometer. 

§ 35. In the tables of Class B, we have the particulars of the series of observations 
made with hydrometers 21 and 17 respectively from which the final mean value of S 
in the table is obtained. In the tables of this class m has the same signification 
as in those of Class A ; S21 gives the mean specific gravity for the particular value 
of m derived from .S21 series of observations made with hydrometer No. 21 ; S17 the 
mean specific gravity similarly obtained from Sy, series made with hydrometer No. 17. 
The final mean derived from s ( = s^i -t- s-^-j), the sum of these series, is found under S. 
Under r^ we have the probable error of S calculated by the method of least squares, 
and under d, the maximum departure of the mean of any individual series from the 
mean specific gravity, S. Numbers under r^ and d are expressed in units of the sixth 
decimal place. 

For each table of Class A referring to specific gravities derived from observations 
with two hydrometers a corresponding table of Class B has been prepared. In the 
twenty-four tables of Class B there are 189 entries under S, r^, and d respectively. 
Summing those under s, we find that the experimental material on which these tables 
are founded consists of 1227 series, whence the mean number of series of observations 
per solution is 6 '49. Each series consists of nine individual observations, and when 
each of them is compared with the corresponding observation made under the same 
conditions in distilled water, they give a mean per solution of 58-4 independent 
observations of the ratio of the weight of a given bulk of the saline solution to the 
same bulk of distilled water, both liquids being at the same temperature. The 1227 
series accounted for in the twenty-four tables correspond to 11,043 independent 
observations of the hydrometric displacement, from each of which the specific gravity 
of the solution in which the instrument floated is deducible. 

The sum of the values of r^ in the twenty-four tables is 5487, which divided by 
189 gives dz2-90, in the sixth decimal place, as the mean probable error of the mean 
specific gravity found for any one of' the solutions. We have seen that this depends 
on a mean of 6-49 series per solution; therefore, admitting that the probable error 



104 MR J. Y. BUCHANAN ON THE 

varies inversely as the square root of the number of observations, the mean probable 
error of the mean specific gravity derived from any number of series s is as follows : — 



s = 1 


2 


3 


4 


5 


6 


7 


8 


9 


±)-„ = 7-39 


5-22 


4-27 


3-69 


3-30 


3-02 


2-79 


2-61 


2-46 



Further, the probable error of the mean of one series being zb7'39, and each series 
consisting of nine individual observations, the probable error of a single observation 
must be 3 x 7"39 = ±22-17 in the sixth place, or ±2*22 in the fifth place. 

§ 36. Following the tables of Class B we have those of Class C, which give a sum- 
mary of the specific gravities of the solutions of diff"erent salts at difi'erent temperatures. 
The salts included in each table have a common acid or a common base. Thus the 
first table of the class contains only chlorides, the second only bromides, and so on. 
These tables furnish the means of comparing the effect of concentration and of the 
specific nature of the salt dissolved on the specific gravity of the solution. 

§ 37. The specific gravity, S, of one of our solutions expresses the weight in kilograms 
of the quantity of the solution having the composition m.MR grams of salt plus 1000 
grams of water, which exactly displaces 1 kilogram of distilled water having the 
temperature T. When we compare the specific gravities of the difi'erent solutions, we 
are considering equal volumes of those solutions ; but the proportion between the salt 
and the water present in this volume of solution is different for different solutions. 
Therefore the specific gravities of the solutions alone do not offer a simple theme for 
discussion. 

The values of W, on the other hand, contain always the constant quantity of water 
in which the different salts are dissolved in quantities proportional to their molecular 
weights. It follows, therefore, that the values of (W— 1000) = w are always exactly 
proportional to the molecular weights of the salts used. 

If the increments of specific gravity (S— 1) were also proportional to the molecular 
weight of the salt dissolved, the quotient W/S = A would be constant for all the solutions 
of the different salts having the same molecular concentration. This is not found to be 
the case. The increments of specific gravity do not follow the periodic law exactly, 
although in the nature of things they cannot depart very far from it. 

But we may consider the specific gravity of a solution from another point of view. 
Let us consider a kilogram of water having the temperature T ; it fills a certain space 
which we may call 1 litre (Lx). We propose to make the solution having the con- 
centration 1/2 KCI4- 1000 grams of water by dissolving portions of the salt KCl in the 
water, but removing so much pure water from the litre-flask as to keep the sum of the 
volumes of water and salt always equal to the litre. When we have in this way 
prepared our litre of (1/2 KCl -f- 1000 grams water), it weighs 1022"98 grams, and is 
composed of 36*78 grams KCl and 986'20 grams of water. As we started with 1000 
grams of water, we have had to remove 13*8 grams of it in order to make room for 
the 36*78 grams KCl which have been dissolved. Consequently, in the construction of 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 105 

the solution we have replaced water by KCl in the proportion of 2-665 grams KCl per 
gram of water. 

We have carried out the calculation for the volume of a litre of initial water and 
final solution. It is much simpler when we take the volume displaced by the weight 
W of the solution, or A = W/S. 

For the solution 1/2 KCl +1000 grams water A =1014-001 Gt. We then take 
1014'00l grams of water, and we add KCl, removing at the same time pure water so as 
to preserve the constant displacement 1014-001 Gtt. We may imagine that equal small 
portions of KCl added take possession of the amount of water required to form with it 
a solution of the concentration (1/2 KCl +1000 grams water), and that the remainder 
of the water is uncontaminated. We proceed on this principle with the fractional 
dissolution of the salt and removal of water so as to keep the displacement constant. 
When we have removed 14001 grams of water, we find that we have dissolved 
37-3 grams or 1/2 KCl in the 1000 grams of water remaining. 

But the operation so described is one of substitution. Consequently it is legiti- 
mate to regard solutions as products of substitution. In fact, the result of the 
operation is that we have replaced 14-001 grams of water by 1/2 KCl, so that the 
substitution has taken place at the rate of 28-002 grams or r555 gram-molecules 
of water per gram-molecule of KCl. 

If we turn to Table No. 82, the first table in Class E, we find the first entry in the 
fourth column is 28-00 as the value oi v/m, or the mean increment of displacement per 
gram-molecule of KCl in the solution (1/2 KC1+ 1000 grams of water) at 19-5° C. 

The tables of Class D give a summary of the Increments of Displacement, v, 
caused by the dissolution of m grm.-mol. of salt in 1000 grams of water at diff'erent 
temperatures. Here 1; = A — 1000. The arrangement of the tables in this class 
is similar to that of Class C. 

The tables in Class E enable us to see at a glance the comparative volumetric 
efi"ect of dissolving different quantities of diff'erent salts in 1000 grams of water. 
Each entry in these tables is derived from the corresponding entry, v, in the corre- 
sponding table of Class D, by increasing it in the proportion w : 1, whence we obtain 
the values v/m. 

§ 38. In the following table we have solutions of the eighteen salts of the 
double ennead (MR, MRO3) for which m = l/16. It gives under w the weight of 
1/16 gram-molecule of the salt dissolved in 1000 grams of water; under S, the specific 
gravity of this solution at 19-5° C, referred to that of distilled water at the same 
temperature as unity ; and under v, the increment of displacement caused by dissolv- 
ing 1/1 G gram-molecule of the salt in 1000 grams of water, expressed in grams of 
distilled water having the temperature 19-5° C. 

The solutions are arranged in three groups, each group containing six solutions of 
salts having the same metallic base (K, Rb, or Cs). These six solutions fall into two 
groups of three, or triads, the first three being the salts having the general formula MR, 

TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 14 



106 



MR J. Y. BUCHANAN ON THE 



and the second triad those having the general formula MRO3. Each triad is entered 
in the ascending order of the molecular weights of the salts which compose it, and each 
group of three triads forms the ennead MR or MRO3 respectively. 



Salt ill Solution. 


Molecular 
Weight of Salt. 


Weight of 

1/16 grm.-mol. 

Salt. 

to. 


Specific Gravity of Solution 

of 1/16 grm.-mol. Salt in 

1000 grams Water. 

S. 


Increment of Displacement 
produced by the Dis- 
solution of w grams Salt 
in 1000 grams Water. 

V. 


KCl 


74-6 


4-6625 


1002973 


1-684 


KBr 


119-1 


7-4437 


1-005279 


2-153 


KI 


1661 


10-3812 


1-007588 


2-772 


KCIO3 


122-6 


7-6625 


1-004863 


2-785 


KBrOg 


167-1 


10-4443 


1 007662 


2-761 


KIO3 


214-1 


13-3812 


1-011169 


2-189 


RbCl 


121-0 


7-5625 


1-005531 


2-020 


RbBr 


165-5 


10-3437 


1-007868 


2-456 


Rbl 


212-5 


13-2812 


1-010046 


3-203 


RbClOj 


169-0 


10-5593 


1-007354 


3-183 


RbBrOg 


213-5 


13-3412 


1-010253 


3-057 


RblOg 


260-5 


16-2812 


1-013673 


2-576 


CsCl 


168-5 


10-5312 


1-008036 


2-475 


CsBr 


213-0 


13-3125 


1-010409 


2-873 


Csl 


260-0 


16-2500 


1-012529 


3-675 


CsClOg 


216-5 


13-5312 


1-009825 


3-669 


CsBrOg 


261-0 


16-3125 


1-012756 


3-511 


CSIO3 


308-0 


19-2500 


1-016299 


2-903 



If we consider the increments of displacement, v, produced by the dissolution of 
1/16 gram-molecule of each of these salts in 1000 grams of water, we see that, for the 
salts of the same metal, the values increase from the chloride to the bromide, and from 
the bromide to the iodide ; and that for the chlorates the values of v are almost identical 
with those for the iodides ; they diminish from the chlorates to the bromates, and suffer 
a considerable fall from the bromates to the iodates. There is also a decided fall in the 
value of V from that of the iodate of potassium or rubidium to that of the chloride 
of rubidium or caesium respectively. 

The following diagram, illustrative of the above table, shows graphically, by the 
heights of the columns, the different increments of displacement produced by the 
dissolution in 1000 grams of water at 19*5° C. of 1/16 gram-molecule of each salt 
of the double ennead (MR, MRO3). The columns representing the increment of 
displacement produced by salts of the ennead MRO3 are shaded. It shows in a very 
striking manner the regular periodic variation of values of v from ennead to ennead. 
It is unfortunate that a complete series of solutions of higher concentration of all the 
salts of the double ennead cannot be obtained, on account of the sparing solubility of 
the oxyhalides, especially those of rubidium and caesium. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 107 



OS 






1-684 



KCl 
74-6 







?,-785 










2-772 












^ 


2-761 








Willi 


^ 








mill!, 


mm 








2-189 




2-153 




^^ 


mim 


iim 


2-020 


^ 


^H 


CD 


r-H 


i-H 


o 


CD 


CO 


(N 


I> 


■* 


^^ 


1— 1 


1» 


(M 


CO 


I— 1 


(M 


T—i 


f-H 


rt 




0^1 


I-H 


ll 


II 


II 


II 


II 


11 






o" 


O 


J, 


_ 


u 






s-< 


o 


o 


CQ 


h-( 


u 


M 


p— 1 


o 


w 


M 


w 


W 


W 


Ph 







' 




3-675 










3-669 
















^: 


3-511 














^^ 


^^ 






3-203 










^ 


^ 






3-183 










^ 


3-057 








iiiiiiiim 


milium 






2-903 




2-873 






^^ 


^^ 








^^ 


^^ 


^^ 






^ 


^ 








^ 


^^ 


^ 






^ 


^ 










^ 


^^ 


^ 




2-576 








""^^^^T"^ 


"--^l!!!;;;^;*-^ 


"---..,^^ 








"•-CI;;;;;---^ 


'"■--^^r^-^ 


"■~-«I^||^ 






^ 


^ 


^ 








^ 


^ 


^ 




2-475 


2-456 




1 


miiiiiiiiiiim 


^ 








^ 


1 


1 


ip 


"P 


o 


>o 


in 


lO 


o 


o 


w 


o 


P ~ 


IC 


(?5 


05 


M 


6 


00 


03 


o 


(o 


1— ( 


9S 


CO 


I-H 


CO 




cc 


CO 


T— 1 


CO 




CD 


o 




CM 


r-\ 


CN 


(M 


I—H 


<M 


CM 


(N 


<M 


CO 


ll 


II 


II 


l 


II 


II 


II 


II 


II 


II 


II 




42 


a 




1— 1 


^ 




1 — 1 

02 


O 

O 

CO 




1— I 


P5 


P? 


« 


P5 


Ccl 


O 


o 


o 


Q 


o 


o 



Formulae and Molecular Weights of Salts. 



Section VII. — The Displacement of the Solutions. 



§ 39. When successive equal quantities of a salt are dissolved in a constant quantity 
of water, the successive increments of displacement of the solution, so produced, are 
generally unequal. They are usually the greater, the greater is the amount of salt 
which has already been dissolved. 

In order usefully to discuss the change produced by any physical action, it is 
advisable bo compare it with that which would be produced if it acted in accordance 
with some law which can be specified with precision. If the results observed agree 
with those calculated in terms of the law postulated, it is good evidence that the 
particular physical action takes place under the law. If no such agreement appears, 
then the observed results must be compared with those calculated in terms of the 
specification of some other law. 



108 MR J. Y. BUCHANAN ON THE 

For the purpose of discussing the changes of displacement produced in a constant 
quantity of water by the dissolution of successive quantities of a salt in it, we compare 
them with those which would take place under one of two hypotheses. 

§ 40. First Hypothesis. — It is assumed that, when a quantity of salt, insufficient for 
saturation, is dissolved in a quantity of water, it takes possession of the quantity of 
water which it requires in order to produce a saturated solution, and saturates it, after 
which the saturated solution disseminates itself through the remaining water, forming 
a simple mixture with it. 

To take a particular case : — Let the constant quantity of water be 1 kilogram, 
and, when saturated with the particular salt used, let it take up 4 gram-molecules 
of it. When 1 gram-molecule of the salt has been dissolved in it, let the displace- 
ment of the solution so produced be r030 kilogram. We have then one-fourth of the 
water saturated by the first gram-molecule of salt added, producing an increment of 
displacement amounting to 30 grams. Let us now add a second gram-molecule of 
salt. There are 750 grams of free water remaining, and of these the second gram- 
molecule of salt takes possession of 250 grams, with which it forms a saturated 
solution ; and this, with the 250 grams saturated water already present, disseminates 
itself through the remaining 500 grams of free water, and forms a homogeneous mixture 
of the two liquids. 

If the increment of displacement produced by the dissolution of the first gram- 
molecule was 30 grams, that produced by the dissolution of the second must be the 
same, because it has been produced by an exact repetition of the first operation ; and 
the total displacement after addition of the second gram-molecule salt must be 
1"060 kilogram. Similarly, when the third gram-molecule has been dissolved, the 
total displacement will be 1'090 kilogram, and, when the fourth gram-molecule 
has been added, and saturation has been reached, the displacement must be 
ri20 kilogram. 

The numerical criterion, therefore, by which to decide if the aqueous solution 
of a particular salt follows this law is that, for equal additions of salt dissolved in a 
constant quantity of water, equal increments of displacement are produced. 

If, by A, we represent the displacement of the solution produced by dissolving 
n parts of the salt in a constant quantity of water, then the above criterion finds 
expression in the equation : 

— = Const. 
dn 

^41. Second Hypothesis. — It is assumed that, when a quantity of salt, insufficient 
to produce saturation, is dissolved in a quantity of water, it exercises no selection, 
but salinifies every particle of the water alike, producing a homogeneous solution of 
uniform concentration, and that, when a second quantity of salt, equal to the first, 
is dissolved in this solution, it intensifies its salinity uniformly and produces an 
increased displacement, which bears the same proportion to that of the first solution 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 109 

as the displacement of the first solution bore to that of the original quantity of water ; 
further, that when a third, equal, quantity of salt is added to the solution of the second 
quantity, it intensifies its salinity uniformly and produces an increased displacement, 
which bears the same relation to that of the second solution as the displacement of 
the second solution bore to that of the first, and as that of the first bore to that of 
the original water ; and so on. 

As we may consider the equal quantities of salt successively added to the constant 
quantity of water to be as small as we please, our imagined process of solution becomes 
more and more nearly continuous. 

If we represent by \ the displacement of the constant quantity of water, and by 
r the ratio of its displacement to A^, that of the solution produced by the dissolution 

of the first of a series of n equal quantities of salt, then '^= ^ and is the common ratio 

of the displacement of each solution to that of the succeeding one in the series. 
Then the displacement of the solution after the n*^ portion of salt has been dissolved 
is expressed by the equation : 

This equation expresses the fact that, when the quantities of salt dissolved in a 
constant quantity of water form an arithmetical series, the displacements of the 
respective solutions so produced form a geometrical series ; consequently the logarithms 
of these displacements form an arithmetical series. 

A convenient numerical criterion, therefore, by which to decide if the 
aqueous solution of a particular salt follows this law, is furnished by the degree of 
conformity of the observed displacements with those calculated on the basis of the 
equation : 

1M^ = Const. 
dn 

To take an example : — As before, let the constant quantity of water be 1 
kilogram,^ which becomes saturated when 4 gram-molecules of the salt used are 
dissolved in it. When the first gram-molecule has been dissolved in it, let the 
displacement of the solution so produced be 1030 kilogram. As the displacement of 
the water was TOGO kilogram, the effect of the first operation has been to increase 
the total displacement in the proportion 1000 : 1 '030 = 1*030. When the second 
gram-molecule of salt is dissolved, it, by hypothesis, increases the displacement of the 
solution containing the first gram-molecule in the same proportion as the dissolution 
of the first gram-molecule increased that of the water, that is, in the proportion 
rOOO : 1*030 ; consequently the displacement of the solution containing the first two 
gram-molecules of salt must be (1*030)1 Similarly, when the third gram-molecule 
has been dissolved, the total displacement must be (1*030)^; and, when the fourth 
gram-molecule has been dissolved and saturation has been produced, the displacement 
must be (1*030)\ 



110 



MR J. Y. BUCHANAN ON THE 



§ 42. The following table gives the means of comparing the effect produced by- 
diluting or concentrating a given solution according as it takes place in terms of the 
first or the second of these hypotheses. The hypothetical salt MR has a molecular 
weight 160, which is nearly the mean of those of the ennead MR. The fundamental 
solution, from which all the others included in the table are derived, is that for which 
m= 1/2, and it is made by dissolving 80 grams of the salt in 1 kilogram of water. 
It is assumed that the dissolution of this mass of the salt in the kilogram of water at 
the standard temperature T, which may be 19 "5° or any other, increases its displacement 
by 20 grams, so that the displacement of the fundamental solution is r020 kilogram. 
The concentration of this solution is then supposed to be altered by the addition to, 
or the withdrawal from, the kilogram of water, of salt so as to produce the series of 
solutions having the concentrations indicated under m in the first column of the table. 
From this the weight of each solution, W, given in the second column, follows as a 
matter of course. 

Hypothetical Case. 
Table giving the calculated Specific Gravities and Displacements of Solutions of 
m.MR+ 1 kilogram of Water at T, where MR= 160 and the Displacement for 
m- 1/2 is 1-020 kilogram of Distilled Water at T. 



m. 


W. 


Sa. 


l-000 + 0-04m 
= Aa. 


Log A]/2X 2m 


Al. 


Si.. 


Sa-S,. 


10 


2 


600000 


1-857142 


1-400000 


0-1720034 


1-485947 


1-749725 


0-107417 


9 


2 


440000 




794118 




360000 


0-1548030 


1-428346 


1-708390 


85728 


8 


2 


280000 




727273 




320000 


0-1376027 


1-372785 


1-660856 


56417 


7 


2 


120000 




656250 




280000 


0-1204023 


1-319478 


1-606695 


49555 


6 




960000 




580645 




240000 


0-1032020 


1-268242 


1-545447 


35198 


5 




800000 




500000 




200000 


0-0860017 


1-218994 


1-476627 


23373 


4 




640000 




413792 




160000 


0-0688013 


1-171659 


1-399724 


14068 


3 




480000 




321368 




120000 


00516010 


1-126162 


1-314197 


7171 


2 




320000 




222222 




080000 


0-0344006 


1-082342 


1-219473 


2749 


1 




160000 




115384 




040000 


0-0172003 


1-040400 


1-114955 


329 


1/2 




080000 




058823 




020000 


0-0086001 


1-020000 


1-058823 


0-0 


1/4 




040000 




029703 




010000 


0-0043000 


1-009951 


1-029753 


-50 


1/8 




020000 




014925 




005000 


00021500 


1-004963 


1-014963 


-38 


1/16 




010000 




007481 




002500 


00010750 


1-002478 


1-007503 


-21 


1/32 




005000 




003745 




001250 


0-0005375 


1-001238 


1-003757 


-12 


1/64 




002.500 




001874 




000625 


0-0002687 


1-000619 


1-001880 


-6 


1/128 




001250 




000937 




000313 


0-0001343 


1-000309 


1-000940 


-3 


1/256 




000625 




000468 




000156 


0-0000671 


1-000155 


1-000470 


_ 2 


1/512 


1-000313 


1-000234 


1-000078 


0-0000335 


1-000077 


1-000235 


-1 


1 


2 


3 


4 


5 


6 


7 


8 



We now apply our first hypothesis (§ 40), and by means of it we obtain the displace- 
ment A^ of each of these solutions expressed in terms of the kilogram of distilled water 
of the standard temperature as unit, and recorded in column 4 of the table. Their 
specific gravities, the quotients obtained by dividing the weight of each solution 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. Ill 

(column 2) by that of the distilled water which it displaces (column 4), are entered in 
column 3 of the table under Sa- 

Let us now apply the second hypothesis, as specified in § 41. The fundamental 
solution is that for which m = 1/2 and if = 80 grams, the displacement of which is taken 
as 1-020 kilogram ; its logarithm is therefore 0-0086001. 

The log-displacement of the solution for which m = 1 is then 

log Aj = 2 X 0086001 = 0-0172002, 
and the log-displacements entered in column 5 under A^ are obtained from the equation 

logA,„ = 0-0172002xm. 

The values of the displacements corresponding to these logarithms are given in 
column 6 under A^. The quotient obtained by dividing the weight of the solution 
(column 2) by the corresponding displacement (column 6) gives the specific gravity of 

the solution, -r- = S^, the values of which are given in column 7. 
Al 

For the value of m= 1/2 the numbers in columns 3 and 7, and in columns 4 and 6, 
are of course identical. For values of m greater than 1/2 the values of Aj^ are always 
greater than those of A^, and for values of w less than 1/2 they are less ; but, as the 
value of m increases the difference A^— A^ increases, so that when m = 10, A^— Aj^^ = 
0-085947. When the value of m diminishes, the difference A^ — A^ diminishes also. 
If we turn to the specific gravities, we see that, in the case where m= 1/32, S^ — &£ = 
— 0-000012, so that, at this concentration, it is still possible to determine by observa- 
tion whether the change of displacement with change of concentration has taken place 
according to the terms of the first or the second hypothesis. For greater dilutions the 
differences of specific gravity approach too nearly to the probable uncertainty of the 
observations to make this possible. 

§ 43. If we study the tables in this memoir, we shall find that in the solutions of 
the majority of the salts the values of dAjdni and vjm reach a minimum for values of m 
in the vicinity of 1/32, and they increase whether the concentration of the solution is 
increased or diminished. At concentrations corresponding to m>2 a number of the salts 
give solutions which conform nearly to the arithmetic law of the first hypothesis. The 
salts which furnish solutions which conform most closely to this law are those which 
contain at least one of the elements Li, Cs, or I. In the ennead MR, after the caesium 
salts and the iodides come some of the rubidium salts and the bromides ; the remainder 
of the bromides and nearly all the chlorides conform more nearly to the geometric law 
of the second hypothesis, and some of them may be said to conform exactly to it. 

§ 44. From the equation log A,„- 0-0172002 x m it follows that A„^ = (1-0404)"'. 
Therefore, if the solutions of a salt follow strictly the law expressed by our second 
hypothesis, the general expression for the displacement of a solution containing m.MR 
in 1 kilogram of water, when the displacement for any particular value of m — for 
instance, for m = 1 — is Aj, is A^ = A^"*. 



112 



MR J. Y. BUCHANAN ON THE 



When the solution does not follow this law quite exactly, let the displacement for 
any particular value of m be A^ = Ai ; then the degree in which the solution conforms 
to the law is indicated by the difference x —m. 

In the table, the displacements given in column 6 are calculated on the basis of 
the second hypothesis. For them, therefore, the relation A^ = A^™ holds good, and the 
value of m (column 1) for any solution expresses not only its molecular concentration, 
but also the exponent of its displacement, that of A^ being taken as unity. 

The values of the displacement of the solutions in column 4 of the table are arrived 
at on the basis of our first hypothesis ; consequently any value of A in this column is 
given by the equation A,,^ = 1 -000 + 0"04m. But none of the solutions dealt with in this 
memoir follow this law at all concentrations, though some of them approximate to it at 
high concentrations. It is therefore of use, in order to augment the illustrative value 
of the table, to determine the exponents of the values of A in column 4 when referred 
to that of A = 1 -020 for m = 1/2 as 1/2. This has been done, and the results are entered 
in the following table : — 



m. 


X. 


x-m. 


m. 

1/2 


X. 


l/m-1/x. 


10 


8-495 


- 1-505 


1/2 


0-00 


9 


7-76 


-1-24 


1/4 


1/3-98 


+ 0-02 


8 


7-00 


-1-00 


1/8 


1/7-943 


+ 0-057 


7 


6-23 


-0-77 


1/16 


1/15-873 


+ 0-127 


6 


5-43 


-0-57 


1/32 


1/31-706 


+ 0-294 ■ 


5 


4-60 


-0-40 


1/64 


1/63-412 


+ 0-588 


4 


3-75 


-0-25 


1/128 


1/127-52 


+ 0-480 


3 


2-86 


-0-14 


1/256 


1/254-06 


+ 1-940 


2 


1-94 


-0 06 


1/512 


1/508-90 


+ 3-100 


1 


0-99 


-0-01 








1/2 


0-50 


000 









§ 45. In the following tables the displacements of most of the solutions have been 
treated along these lines. In the first four tables we have for the solutions of the salts 
of the ennead MR the values of x and of x—m for the strong solutions, and those 
of X and 1/m - l/x for the dilute solutions. For solutions of other salts the tables give 
only the values of x —m or 1/m - l/x, as they are sufficient. 

The numbers representing the values of x and m for the strong solutions are the 
numerators of vulgar fractions having unity for common denominator. The numbers 
representing the values of l/x and 1/m for the weak solutions are the denominators of 
vulgar fractions having unity for common numerator. The measure of the departure of 
the displacements of solutions of a particular salt from the geometric law of the second 
hypothesis is found for the strong solutions in the column headed {x — m). For the weak 
solutions the corresponding column is headed (l/m — l/x), so that the signs prefixed to 
the numbers in these columns mean the same thing in both tables : — the + sign means 
that x>w, the — sign that x<jn. In the strong solutions x = m when m= 1, and the 
remaining values of m increase; in the weak solutions x = m when m=l/2, and the 
remaining values of m diminish. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 113 



Table I. 
Values of X for the Ennead ME. {Strong Solutions.) 









T = 19-5°C. 








T = 21-4°C. 


T=23-rC. 


m. 


KCl. 


KBr.* 


KI. 


RbCl. * 


RbBr. 


Rbl. 


CsCl. 


CsBr. 


Csl. 


1 


1-00 


1-00 


1-00 


1-00 


1-00 


100 


1-00 


1-00 


1-00 


2 


2-04 


2-01 


1-98 


2-01 


1-97 


1-97 


1-97 


1-96 


1-96 


3 


3-09 


3-01 


2-92 


3-03 


2-96 


2-92 


2-94 


2-95 


2-86 


4 


413 


4-00 


3-86 


4-00 


3-93 


3-82 


3-92 


3-84 


3-76 


5 




4-95 


4-76 


4-95 


4-86 


4-69 


4-80 


4-81 




6 






5-63 


5-93 


5-78 


5-53 


5-74 






7 






6-48 


6-88 




6-41 


6-58 






8 






7-30 








7-39 






9 


i 












8-24 







* Compare table, p. 173. 

Table II. 
Values of X - m for the Ennead MR. [Strong Solutions.) 









T = 


19-5° 0. 








T = 21-4°C. 


T = 23-1°C. 


m. 


KG]. 


KBr. 


KL 


RbCl. 


RbBr. 


RbL 


CsCl. 


CsBr. 


CsL 


1 


000 


0-00 


00 


0-00 


0-00 


0-00 


0-00 


0-00 


000 


2 


0-04 


0-01 


-0-02 


0-01 


-0-03 


-0-03 


-003 


-0 04 


-0-04 


3 


0-09 


0-01 


-0-08 


0-03 


-0-04 


-0-08 


-0-06 


-005 


-0-14 


4 


013 


0-00 


-0-14 


000 


-0-07 


-0-18 


-0-08 


-0-16 


-0-24 


5 




-0-05 


-0-24 


-0-05 


-0-14 


-0-31 


-0-20 


-0-19 




6 






-0-37 


-0-07 


-0-22 


-0-47 


-0-26 






7 






-0-52 


-0-12 




-0-59 


-0-42 






8 






-0-70 








-0-61 






9 














-0-76 







Table HI. 

Values of x for the Ennead MR. 

T = 19-5°C. 



1/2 
1/4 
18 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/1024 



KCl. 



1/2 

1/4-04 

1/8-14 

1/16-54 

1/33-06 

1/65-60 

1/127-52 

1/282-24 

1/434-42 



KBr. 



1/2 

1/4-02 

1/8-08 

1/16-16 

1/32-18 

1/62-82 

1/124-90 

1/249-02 

1/466-90 



KL 



1/2 

1/4-00 

1/8-08 

1/16-26 

1/32-30 

1/64-82 

1/129-70 

1/266-98 

1/505-40 

1/1120-46 



RbCl. 



1/2 

1/4-04 

1/8-14 

1/16-34 

1/32-80 

1/67-40 

1/136-22 

1/268-58 

1/450-84 



RbBr. 



1/2 

1/4-04 

1/8-16 

1/16-34 

1/32-84 

1/63-98 

1/130-12 

1/224-98 

1/444-92 



Rbl. 


1/2 

1/3-98 
1/7-94 
1/15-92 

1/31-82 


1/62-70 


1/120-76 


1/233-80 


1/356-20 


l/835'36 



CsCl. 



1/2 

1/3-60 

1/8-12 

1/16-32 

1/32-96 

1/66-80 

1/138-82 

1/279-86 

1/506-40 



CsBr. 



1/2 

1/4-04 

1/8-14 

1/16-36 

1/32-86 

1/64-00 

1/130-28 

1/212-28 

1/440-20 

1/489-28 



CsL 



1/2 

1/3-92 

1/7-98 

1/13-88 

1/32-28 

1/62-32 

1/120-88 

1/211-20 

1/248-94 

1/382-38 



TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 



1.5 



114 



MR J. y. BUCHANAN ON THE 



Table IV. 
Values of l/ni— Ifx for the Ennead MR. 











T 


= 19-5°C. 










m. 


KCl. 


KBr. 


KI. 


RbCl. 


RbBr. 


Rbl. 


CsCl. 


CsBr. 


Csl. 


1/2 


0-00 


000 


0-00 


0-00 


00 


0-00 


0-00 


0-00 


0-00 


1/4 


-0-04 


-0-02 


0-00 


-004 


-0-04 


+ 0-02 


+ 0-40 


-0-04 


+ 0-08 


1/8 


-014 


-0-08 


-0-08 


-0-14 


-016 


+ 0-06 


-0-12 


-014 


+ 0-02 


1/16 


-0-54 


-016 


-0-26 


-0-34 


-0-34 


+ 0-08 


-0-32 


-0-36 


+ 2-12 


1/32 


-1-06 


-0-18 


-0-26 


-0-80 


-0-84 


+ 0-18 


-0-96 


-0-86 


-0-28 


1/64 


-1-60 


+ 1-18 


-0-82 


-3-40 


+ 0-02 


+ 1-30 


-2-80 


0-00 


+ 1-68 


1/128 


+ 0-28 


+ 3-10 


- 1-70 


-8-22 


-2-12 


+ 7-24 


10-82 


-2-28 


+ 7-12 


1/256 


-26-24 


+ 6-98 


- 10-98 


-12-58 


+ 3102 


+ 22-20 


-23-86 


+ 33-72 


+ 44-80 


1/512 


+ 77-58 


+ 4510 


+ 6-60 


+ 5316 


+ 67-08 


+ 155-80 


+ 5-60 


+ 71-80 


+ 263-06 


1/1024 






-96-46 






+ 188-64 




+ 534-72 


+ 641-62 



Table V. 

Values of 1/m— \/xfor the Ennead MRO^. 

T=19-5°C. 



m. 


KCIO3. 


KBrOs. 


KIO3. 


RbClOg. 


RbBrOj. 


RblOj. 


CsClOg. 


CsBiOg. 


CSIO3. 


1/2 




















1/4 


0-00 


0-00 


0-00 


0-00 






000 






1/8 


0-00 


-008 


-0-12 


+ 0-02 






+ 0-04 






1/16 


-0-20 


-0-28 


-0-08 


+ 0-09 


0-00 


0-00 


+ 0-28 


0-00 


0-00 


1/32 


-1-76 


-0-76 


-0-08 


+ 0-05 


+ 0-32 


-0-16 


+ 0-04 


+ 0-32 


+ 0-64 


1/64 


-4-24 


-1-24 


+ 3-76 


-1-24 


+ 0-48 


+ 0-32 


+ 4-64 


-0-80 


+ 5-12 


1/128 


- 11-12 


-1-36 


-2-56 


+ 1-76 


+ 8-00 


+ 9-20 


-2-92 


-0-96 


+ 26-72 


1/256 


-28-28 


+ 1-72 


-20-36 


+ 4-16 


+ 1-92 


+ 30-16 


-59-16 


+ 17 60 


+ 86-08 


1/512 


-274-92 


+ 8-80 


-103-20 


+ 70-08 


+ 6-72 


-53-44 


+ 240-96 


+ 95-04 


+ 206-88 



Table VI. 

Values of 1/m - llxfor Solutions of the Nitrates. 



M"= 1 ^"• 


K. 


Sr". 


Ba". 


Li. 


Na. 


K. 


Rb. 


Cs. 


Ba". 


Pb". 


Rb. 


Cs. 


T = 


15-0° 0. 


19-5° 0. 


23 


■o°c. 


1/2 




























14 










-0-04 


-0-04 


-0-02 


-0-02| 










1/8 










-0-06 


-0-16 


-0-14 


-0-08 


000 






-0-24 




1/16 




-0-14 






-0-38 


- 0-54 


-0-42 


-0-72 


-0-16 






-0-72 


-0-1(5 


1/32 


+ 0-48 


+ 0-02 






-1-32 


-1-28 


-1-60 


- 2-84 


-1-56 


-0-80 


-0 80 


-0-28 


-0-48 


jl/64 +0-72 


+ 0-34 


+ 0-32 


-1-60 


-4 04 


-2-44 


-2-90 


-14-70 


+ 2-88 


-3-04 


-4-48 


-12-32 


-2-56 


1/128 +0-80 


+ 1-90 


-1-44 


- 28-32 


- 13-94 




-4-86 


-59-36 


-5-80 


-16-64 


-7-84 


-93-24 


-21-60 


' 1/256 ' 




+ 4-48 


+ 7-68 








-112-22 


-64-16 


- 35-20 


- 19-84 


+ 200 


- 20-96 


1/512 




+ 23-68 


-6-72 








+ 233-04 


-317-72 


- 109-12 


- 139-84 




- 253-76 


1/1024 






- 228-80 










+ 16156 


+ 69-60 


-951-04 




-3258-32 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 115 



Table VII. 

The following Table gives Values of m and x-m for the Solutions of a nuynher of 
Salts derived from Determinations of their Specific Gravity at 19"5° C. by 
Kremers. {Pogg, 1855, vols. xcv. and xcvi.) 





NaNOj. 


NaClOs. 


NaBrO^. 


LiCl. 


LiBr. 


Lil. 


NaCl. 


NaBr. 


Nal. 


A J for m = 1 


1029-70 


1037-025 


1036-51 


1018-52 


1028-35 


1036-09 


1018-25 


1025-225 


1036-03 


in. 










Values of X - 


m. 








1 























2 


0-U24 


0021 


0-027 


0-056 


0000 


0-000 


0-072 


0037 


-0010 


3 


0-066 


0030 




0-117 


0-067 


-0-032 


0167 


0-089 


-0-035 


4 


0-110 


0011 




0-182 


0-203 


-0-064 


0310 


0-140 


-0-145 


5 


0114 


0-301 




0-232 


0-338 


-0-129 


0-458 


0-186 


-0140 


6 


0-170 






0-271 


0-350 


- 0-265 


0-611 


0-216 


-0-233 


7 


0-177 






0-290 


0-234 


-0-365 




0-234 


-0-340 


8 


0-164 






0-290 


0-125 


-0-585 




0-243 


-0-466 


9 


0131 






0-290 


0-013 


-0-805 




0-240 


-0-635 


10 


0-081 






0290 


-0-190 


-1-025 






-0-803 



In the tables given above we have, under the values of x, the values of log A,„/log A^, 
and they show, when compared with the values of m, the degree in which the displace- 
ment of the solutions follows the logarithmic law. But there is a certain disadvantage 
in taking as our unit of comparison the logarithm of the displacement for m=l when 
the difference between the values of m and 1 is considerable, inasmuch as the effects due 
to departures from the logarithmic law are cumulative. In the following tables we 
have the values of log A,„/log A^ for a number of pairs of solutions of salts belonging 

to the enneads MR and MRO3, and of certain nitrates. 

Table VIII. 

Values of , ^ " for the Ennead MR. 
•^ log ^J 











T=] 


L9-5° C. 










m. 


KCl. 


KBr. 


KL 


RbCl. 


RbBr. 


RbL 


CsCl. 


CsBr. 


Csl. 


8 






1-891 


1-966 




1-894 


1-883 






4 


2-027 


1-992 


1-939 


2-021 


1-977 


1-940 


1-988 


i-oio 


1-981 


2 


2-041 


2-010 


1-990 


2-076 


1-988 


1-972 


1-976 


1-920 


1-969 


1 


2-015 


2-002 


2-006 


2-155 


2-020 


1-991 


2-054 


2-040 


2-107 


1/2 


2-022 


2-010 


2-008 


2-020 


2019 


1-998 


2-016 


2-000 


1-992 


1/4 


2-016 


2-010 


2-018 


2-017 


2-015 


1-992 


2-012 


2-020 


2 006 


1/8 


2-026 


1-999 


2008 


2006 


2-009 


2-002 


2-012 


2-016 


1-995 


1/16 


2-001 


1-988 


1-985 


2-006 


2-008 


1-997 


2-016 


1-958 


2-023 


1/32 


1-983 


1-947 


2-006 


2-053 


1-945 


1-969 


2-021 


2-108 


1-930 


1/64 


1-945 


1-979 


1-999 


2-050 


2-031 


1-923 


2067 


1-765 


1-938 


1/128 


2-202 


1-973 


2-053 


1-937 


1-628 


1-936 


1-994 


1-748 


1-744 


1/256 


1-530 


1-842 


1-882 


1-670 


2-089 


1-521 


1-778 


2-074 


1-176 


1/512 






2-184 




1-091 


2-303 




1-704 


1-529 



116 



MK J. Y. BUCHANAN ON THE 



Table IX. 



Values of^^^-^ for the Ennead MRO.,. 
^ log ^J 











T = 


19-5° C. 










m. 


KCIO3. 


KBiOs. 


KIO3. 


RbOlOg. 


RbBiOj. 


RblOj. 


CSCIO3. 


CsBrOg. 


CSIO3. 


1/4 


2-010 


2-019 


2-031 


1-997 






1-999 






1/8 


2-019 


2-017 


1-980 


1-993 






1-968 






1/16 


2-081 


2-013 


1-996 


2-007 


1-983 


2-018 


2-029 


1-985 


2028 


1/32 


2-021 


1-990 


1-875 


1-982 


2-008 


1-931 


1-859 


2 044 


1-918 


1/64 


2-037 


1-983 


2-165 


1-882 


1-884 


1-924 


2-039 


2-047 


1-788 


1/128 


2-039 


1-965 


1-112 


1-887 


2-122 


1-814 


1-624 


1-792 


1-795 


1/256 


2-746 


1-977 


1-211 


1-653 


1-974 


2-635 


1-376 


1-746 


2-080 


1/512 





















Table X. 



Values of , ^ ^"^ for some Nitrates. 
^ log ^J 



m. 


Li. 


Na. 


K. 


Rb. 


Cs. 


Ba". 


Pb". 


Rb. 


Cs. 


T = 


19-6° 0. 


23-0° C. 


8 

4 

2 

1 
1/2 
1/4 
1/8 
1/16 
1/32 
1/64 
1/128 
1/256 
1/512 


1-939 
1-982 
2-003 
1-984 
2-023 
1-995 
2-030 
2034 
2-040 
2-083 


2-019 
1-993 
2-049 
2027 
2-026 
2-016 
2-020 
2-013 
1-995 


2-014 
2-025 
2-015 
2-046 
1-983 
2-050 


2-000 
2 006 
2-008 
2-015 
2-068 
2-082 
2-258 
2-379 
1-961 


2-006 
2-917 
2-075 
1-880 
2-117 
2-388 


2-048 
2-046 
2-159 
2-010 
2-129 
1-528 


2-056 
2-086 
1-981 
2-031 
2-359 
3-010 


2069 
2-019 
2-091 
2-182 
2-901 
1-146 


2-019 
2-013 
2-048 
2-245 
1-851 



Section VIII. — ^Comments on the Changes in the Values of dA — v for Different 
Values of m in the Case of Solutions op Individual Salts of the Type 

MR AND MRO3. 

§ 46. The earlier introductory notes and discussion of the values of the specific 
gravity and displacement explain the precise meaning of these terms. A comparison of 
the values of the increment of displacement (v) caused by the dissolution of m gram- 
molecules of a salt in 1000 grams of water with the difference (dA) of these consecutive 
increments of displacement possesses considerable interest. 

The increment of displacement (v) due to the dissolution of m gram-molecules of a 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 117 

salt in 1000 grams of water may be looked on as being the result of two operations, 
namely : (a) the dissolution of — gram-molecules of the salt in 1000 grams of water, 
which produces the first increment of displacement ; and (b) the further dissolution of 
-- o'ram-molecules of the salt in the solution formed, which produces the second incre- 
ment of displacement. These increments of displacement are very seldom found to be 
alike ; the second portion of salt dissolved generally produces a greater increment of dis- 
placement than the first. In the case of solutions of such concentration that their specific 
gravity can be ascertained by the use of the pyknometer without too great probability of 
error, this feature is found to be general, and has often been held to be universal. One 
of the principal motives f 07- making this research ivas to find out, by the use of the 
more refined hydrometric method, if there is any point in the dilution of a saline 
solution at which further dilution is accompanied by expansion, in place of contraction. 
The general result of the work is to shoiv that in solutions having the concentrations 
here used, ivhere ?w<l/16, cases of expansion 07i dilution are not uncommon. 

The following table gives the values of m for which the value of (c^A — n) is 
positive and becomes negative for the next lower value of ni. That is, the value 
(^dA — v) changes sign at some concentration lower than that indicated by m and 
higher than that indicated by l/2m. 



T=19-5''C. 


MR 


KCl 


RbCl 


CsCl 


KBr 


RbBr 


CsBr 


KI 


Rbl 


Csl 




m 


1/32 


1/128 


1/256 


1/16 


1/32 


1/64 


1/16 


1/16 


1/32 




MROg 

m 


KCIO3 
1/512. 


RbClOj 
1/64 


CsClOg 
1/32 


KBrOg 

1/32 


RbBr03 

1/64 


CsBiOj 

1/128 


KIO3 

1/16 


Rbl03 

1/32 


CSIO3 
1/16 




MNO3 
m 


LiNO, 

1/2 


NaNOg 
1/32 


KNO3 
1/32 


RbNOg 

1/128 


CsNOg 
1/32 


Ba(N03)2 
1/512 


Pb(N03)2 
1/64 






T = 23-0°C. 


MR 


RbBr 


CsBr 


KI 


Rbl 


Csl 












m 


1/32 


1/8 


1/16 


1/16 


1/32 












MEO3 


RbNOj 


CSXO3 


















ni 


1/128 


1/128 




' 












T=15-0°C. 


MR 
m 

MNO3 
m 


KCl 
1/32 

NaNOg 
1/16 


NaCl 
1/16 

KNO3 
1/16 


Sr(N03)2 
1/128 


Ba(N03)2 
1/128 













§ 47. If we consider Table No. I, Class A, § 26, in which are recorded the results 
of experiment on solutions of chloride of potassium having the concentrations specified 
by the general expression mKCl-t- 1000 grams of water, we have, besides the weight of 
the solution and its specific gravity, columns containing its displacement, A, and the 



118 MR J. Y. BUCHANAN ON THE 

difference, ct?A, between consecutive increasing values of A. If we consider the pair of 
solutions for which m = 1/8 and 1/16, we find that the dissolution of the quantity 1/1 6 KCl 
in 1000 grams of water produces an increment of displacement v= 1*684 Grig.jo. When 
■ we dissolve a further 1/16 KCl in the solution, the displacement is increased by 
dA^ 1-732 Gin-s-. The difference 1732 - 1 "684 = 0-048 is quite genuine. If we now 
consider the pair of solutions for which m= 1/256 and 1/512, we have the displacements 
A = 1000-098 and 1000'064 respectively. Here the dissolution of the first 1/512 KCl 
causes an increment of displacement of 'y = 0064 Gi9-5=, while the dissolution of the 
second 1/512 increases the displacement by only 0-034 G19.5 , showing that contraction has 
taken place when the concentration of the solution has been increased from 1/512 to 2/512 
gram-molecule per 1000 grams of water. From this it follows necessarily that the dilution 
of 1/256 KCl + 1000 grams of water to 1/512 KC1+ 1000 grams of water is accompanied 
by expansion. The specimen table for KCl in § 50 is constructed so as to show the 
character of the change of displacement of a solution with change of its concentration. 
The first line (n) gives the ordinal number of each column ; in the second line (m) we 
have the quantity of salt, expressed in terms of the gram-molecule, dissolved in 1000 
grams of water ; and in the third line (v) we have the increment of displacement in 
terms of grams of water at the temperature of observation, G^. Then follow three 
sub-tables [a, b, c). In sub-table a consecutive values of m, A, etc., are considered; 
in sub -table b alternate values of 771, A, etc., are considered; and in sub-table c values 
are given which represent the mean of those given in a and b. The last line in each 
sub-table contains the values of the differences c?A — v, dA' — v', and dA" — v" , arrived 
at in sub-tables a, b, and c respectively. 

§ 48. Before entering upon a detailed examination of the tables giving values of cZA 
and V and their differences dA — v, we will consider the influence of change in specific 
gravity on the value of the displacement, from which it will be possible to ascertain 
how far the differences dA — v are to be accepted as independent of experimental error. 

For this purpose we will consider the effect of change of specific gravity on the 
least concentrated solution of the salt KCl, the value of m being 1/512, and the 
molecular weight of the salt being 74-6. 

The mean specific gravity is given in the table as 1*000082 ; the weight of soiutioi] 
composed of 1000 grams of water and 1/512 gram-molecule KCl is 1000-1457 grams ; the 
displacement is therefore 1000*064. The difference of displacement between the 1/256 
and 1/512 gram-molecule solution is 0-034. 

We have seen (§ 35) that the mean probable error of the specific gravity of 
any of the solutions entered in the tables. Class A, is ±3 in the sixth decimal place ; and 
this is independent of the concentration of the solution. When the values of the 
displacement are obtained by the use of any of these specific gravities, the probable 
error is only increased by that due to the preparation of the solution, which may be 
neglected (see § 49). The values of displacement so arrived at are affected by a 
probable error which is also independent of the concentration. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 119 



When, however, we consider the values of v, ( = A-1000), or those of cIA, the 
probable error has a close relation to the concentration. Thus, when the value for A is 
1000-064, this effect of an error of ±3 in the third decimal place, which corresponds 
to the sixth place of decimals in the value of the specific gravity, is insensible ; but 
when we consider the value of v = 0-064, a diiference of ±0-003 gives the values 0-061 
and 0-067 as possible values, and this is equivalent to an uncertainty in the value of v 
. having a range of 0-006, which is 94 per cent, of 0-064, the mean value of v. 

If we consider a solution of greater concentration, e.g. that of 1/128 KCl in 1000 grams 
of water, the value of v is 0-217, which, when affected by a probable error of ±0-003, 
gives a possible range of uncertainty of 0-006, and this is only 2-75 per cent, of the 
mean value oft', which is 0-217. 

We have considered the case of KCl, which has a low molecular weight, and its 
solutions have a comparatively low specific gravity. Salts of higher molecular weight, 
such as KI or the salts of rubidium and caesium, form solutions which have higher 
specific gravities for an equivalent concentration. In these cases the range of un- 
certainty of the values of v and d^ is relatively less considerable than that in the 
case of KCl. 

§ 49. The following table furnishes evidence of the possible uncertainty of the 
value of the displacement of a solution which is due to the accumulation of the errors 
affecting the preparation of the solution and those affecting the determination of its 
specific gravity when practised by different experimenters. 

POTASSIUM CHLORIDE. KCl=74-6. 
T = 19-5°C. 



m. 


Weight of 
Solution. 


Specific Gravity obtained by 


Corresponding Displacement. 


D. 


B. 


D. 


B. 


1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 


1009-325 
1004-662 
1002-331 
1001-166 
1000-582 
1000-291 
1000-146 


1-005889 
1-002973 
1-001489 
1-000741 
1-000365 
1-000193 
1-000082 


1-005911 
1-002972 
1-001473 
1-000740 
1-000376 
1-000195 
1-000073 


1003-416 
1001-684 
1000-841 
1000-423 
1000-217 
1000-098 
1000-064 


1003-394 
1001-685 
1000-857 
1000-424 
1000-206 
1000-096 
1000-073 



The numbers in the columns headed D are abstracted from Table No. ] , Class A. 
The experiments were made by Mr H. Royal-Dawson in May 1904, using hydrometers 
Nos. 17 and 21. The scheme for the preparation of the various concentrations of the 
solutions in this series was that of diluting a known quantity of the stronger solution 
with the quantity of water necessary to produce the solution whose strength was one- 
half that of the solution from which it was made ; thus providing a series of solutions 
whose strengths diminish in a geometric succession. 



120 MR J. Y. BUCHANAN ON THE 

The numbers in the columns headed B are abstracted from a series of experi- 
ments made by Mr S. M. Bosworth in March 1911, using hydrometers Nos. 17 and 3. 
The series from which these values have been extracted was an arithmetical one with 
the common difference 1/64 gram-molecule, with the exception of the last three, which 
formed a geometric series. The method of preparation of these solutions was as 
follows : — 

The first solution was made by dissolving 1/64 gram-molecule KCl in 1000 grams of 
water. From this the 2/64 gram-molecule solution was obtained by the addition of 1/64 
gram-molecule KCl to the quantity of the i/64 gram-molecule solution containing 
1000 grams of water. 

Thus in each case the more concentrated solution was prepared from the more dilute 
by the addition of the requisite quantity of salt, while in Mr Royal-Dawson's practice 
the more dilute solution was prepared from the more concentrated by the addition of 
the requisite quantity of water. 

Mr Royal-Dawson's results obtained with solutions of chloride of potassium, quoted 
in the table, were among the first which were accepted as final and admitted into this 
memoir. Before May 1904 Mr Royal-Dawson's work consisted in learning the art of 
the exact use of the hydrometer and in assisting me in the elaboration of the plan 
of making and recording the experiments, which has since been adhered to. Mr 
Bosworth's results were obtained after he had perfected himself in the use of the 
instrument and after the system had been in continuous operation for seven years. 
When it is further taken into account that the two experimenters prepared their 
solutions in almost opposite ways, it will be admitted that the juxtaposition of the two 
sets of experiments subjects the method to a very stringent test, and that their agree- 
ment affords the best evidence of the trustworthiness of the experimental method used 
in this research. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 121 



§ 50. Specimen Table indicating the Stages in the Calculations used in the Discussion 
of the Values of v and c?A in the Case of Solutions of Different CoU' 
centrations of KCl. 



n 


9 


8 


7 


6 


5 


4 


3 


2 


m 


1/4 


1/8 


1/16 


1/32 


1/64 


1/128 


1/256 


1/512 


V 


6-899 


3-416 


1-684 


0-841 


0-423 


0-217 


0-098 


0064 



1/1024 



Sub-table a. 



A -1000 = ^; 

dA -V 



7-102 

6-899 

+ 0-203 


3-483 

3-416 

+ 0-067 


1-732 

1-684 

+ 0048 


0843 

0-841 
+ 0-002 


0-418 

0-423 

-0005 


0-206 

0-217 

-0011 


0-119 

0-098 
+ 0-021 


0-034 

0-064 

- 0-030 



Sub-table b. 



'/„+i - v'„ = dA' 
dA' -v' 





10-585 


5-215 


2-575 


1-261 


0-624 


0-325 


0153 




3-528 


1-738 


0-858 


0-420 


0-208 


0-108 


0-051 




3-522 


1-723 


0-856 


0-418 


0-219 


0091 


0-051 


6-944 


3-422 


1-699 


0-843 


0-425 


0-206 


0-115 


0-064 




+ 0-100 


+ 0-024 


+ 0-013 


-0-007 


+ 0-013 


-0-024 


-0-013 



Sub-table c. 



v"„+i-v"„ = dA" 

v +v „ 

2 ""^ 

dA" -v" 



7-080 


3-502 


1-728 


0-849 


0-416 


0-213 


0-105 


0-042 


6-921 


3-419 


1-691 


0-842 


0-424 


0-211 


0-106 


0-064 


+ 0-159 


+ 0-083 


+ 0-037 


+ 0-007 


-0-008 


+ 0-002 


-0-001 


-0-022 



Note. — The numbers in line w designate the columns of the table ; m is the concentra- 
tion of the solution expressed in gram-molecules of salt dissolved in 1 000 grams of water. 

The significance of the symbols m, c?A, v, and c^A — v has been already explained. 

The differences (c?A - v) are accounted positive when dA is greater than v, and 
negative when the reverse condition holds. 

From an inspection of the values of dA and v in sub-table a, as derived from con- 
secutive values of A, it will be seen that there is a decrease in the positive values of 
dA-v for diminishing values of w from 1/4 to 1/32, when change of sign occurs and 
negative values are obtained for w=l/64 and 1/128, then a reversal to a positive 
quantity of -1-0-021 for 1/256, and again a negative value for 1/512. 

There is thus a steady decrease in the positive values for dA — v from m ^ 1/4 to 
m = 1/32. Between m = 1/32 and m = 1/64 there is change of sign, and from m = 1/64 to 
m= 1/512 there are increasing negative values, with exception of -{- 0-021 at m= 1/256. 

The value -f 0-021 for dA-v at w= 1/256 is of appreciable magnitude, but the 
concentration of the solution is low, and, as shown in the example for calculating the 
effect of specific gravity changes on the displacement, its relative value would be con- 
siderably affected by slight changes in the value of the specific gravity. 

If we now consider alternate values v^+i and v^_i instead of consecutive values 
TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 16 



122 MR J. Y. BUCHANAN ON THE 

Vn and Vn+i, the difference of these quantities as given in sub-table h, line j, includes a 
value for dA which is approximately three times greater than that obtained by taking 
consecutive values v^+i and v^ as in sub-table a. 

Hence one-third of the difference of displacement {line k, sub-table b), added to the 
lower value v„_i employed for the purpose of obtaining the difference, gives a new 
interpolated value v' for the solution of m (see line m, sub-table h). 

From this series of interpolated values of v^ a corresponding set of values, clA' (see 
line I, sub-table b), is obtained by taking the difference of consecutive values of v', 
and the differences of these values, dA' and v', give a new series of values, dA' — v\ set 
out in line n, sub-table b. 

A comparison of the values of dA' — v' (sub-table b) with those of dA — v (in sub- 
table a) shows the same general character of the change of values and of sign from 
m= 1/8 to m= 1/512, only that the reversal of sign takes place at m= 1/128 instead of 
at 771 = 1/256 as in sub-table a. 

A third series of values is derived by using the mean of corresponding values of 
V and v', namely, ^{v + v') = v" , and obtaining the difference dA" — v" in the same way 
as the corresponding values of dA — v and dA' — v' were obtained. 

An inspection of these new values shows that the character of the decrease of the 
positive value is the same as in the other two cases, but the reversal of sign remains at 
7n= 1/128 as in sub-table b. The magnitudes of the values d/\" — v" when w<l/32 
are, however, so small that they would indicate no comparative change of volume wlien 
successive small increments of salt were added to these dilute solutions. 

These calculations serve to indicate the character of the relative changes of volume 
occurring when successive dilutions are effected. 

The results of the treatment of the values of v in the case of KCl justifies a similar 
treatment of the values of v in the solutions of the other salts included in the series of 
tables. Class A, and the results for the salts of the types MR and MROg are given in 
§ 5 1 in a series of tables of Class F for each salt, similar to the specimen table given 
above, and arranged in triads of the salts having the same acid radical, R or ROg as the 
case may be. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 123 



5L Tables of Class F, illustrating the Method of arriving at the Volumetric Effect 
produced by changing the Concentration of a Solution. 

TRIAD OF CHLORIDES. 

Table I. POTASSIUM CHLORIDE. KCl = 74-6. T = 19-5° C. 



9 

1/4 
6-899 



1/8 
3-416 



7 
1/16 
1-684 



6 
1/32 
0-841 



5 
1/64 
0-423 



1/128 
0-217 



3 

1/256 
0-098 



2 

1/512 
0-064 



1 

1/1024 



Sub-table a. 



A-1000-«; 
d^ -- V 



7-102 

6-899 

+ 0-203 



3-483 
3-416 

+ 0-067 



1-732 

1-684 

+ 0-048 


0-843 

0-841 

+ 0-002 


0-418 

0-423 

-0-005 


0-206 

0-217 

- 0-011 


0-119 

0-098 

+ 0-021 


0-034 

0-064 

-0-030 



Sub-table b. 



v„+i - V„-i 

v'n+\-v'n = dA' 
l-'n-l + kiVn+l-Vn-l)-- 
clA' - v' 



6-944 



10 


585 


3 


528 


3 


522 


3 


422 


+ 


100 



5-215 


2-575 


1-261 


0-624 


0-325 


0-153 


1-738 


0-858 


0-420 


0-208 


0-108 


0-051 


1-723 


0-856 


0-418 


0-219 


0-091 


0-051 


1-699 


0-843 


0-425 


0-206 


0-115 


0064 


+ 0-024 


+ 0-013 


-0-007 


+ 0-013 


-0-024 


-0-013 



SUB-TABLH C. 



v"„+i-v"n = dA" 

dA"-v" 



7-080 

6-921 

+ 0-159 



3-502 

3-419 

+ 0-083 



1-728 

1-691 

+ 0-037 



0-849 


0-416 


0-213 


0-105 


0-042 


0-842 


0-424 


0-211 


0-106 


0-064 


+ 0-007 


-0-008 


+ 0-002 


-0-001 


-0-022 



Table IL RUBIDIUM CHLORIDE. RbCl = 121-0. T=19-5° C. 

Sub-table a. 



Vn+l - V„ = dA 

A-1000 = r 

dA - V 



8-435 

8-202 

+ 0-233 



4-145 

4-057 

+ 0-088 



2-037 

2-0-20 

+ 0-017 



1-014 

1-006 

+ 0-008 



0-517 

0-489 

+ 0-028 



0-251 

0-238 

+ 0-013 



0-116 

0-122 

-0 006 



0-049 

0-073 

•0-024 



Sub-table b. 



Vn+i - Vn~i 

Uv„+-i - V„-i) 

v'n+i-v'n = dA' 

dA' -v' 



8-250 



12-580 
4-193 
4-170 
4-080 



6-182 
2-060 
2-057 
2-023 



3-051 
1-017 
1-024 
0-999 



1-531 
0-510 
0-505 
0-494 



+ 0-090 +0-034 +0-025 +0-011 +0-006 -0-066 +0-009 



0-768 
0-256 
0-250 

0-244 



367 

122 
089 



0-155 



0-165 
0-082 
0-082 
0-073 



Sub-table c. 



v"„ + i-v"n = dA" 

2 
dA" - v" 



8-411 

8-226 

+ 0-185 



4-158 

4-068 

+ 0-090 



2-047 

2-021 

+ 0-026 



1-019 

1-002 

+ 0-017 



0-511 

0-491 

+ 0-020 



0-250 

0-241 
+ 0-009 



0-103 
0-138 
035 



0-065 

0-073 

-0-008 



Table in. CiESIUM CHLORIDE. CsCl = 168-5. T = 19-5°C. 

Sub-table a. 



VH+l-Vn = dA 

A-1000 = 'u 
dA -V 



10-335 

10066 

+ 0-269 



5-077 

4-989 

+ 0-088 



2-514 

2-475 

+ 0-039 


1-250 

1-225 

+ 0-025 


0-621 

0-604 

+ 0-017 


0-313 

0-291 
+ 0-022 


0-147 

0-144 

+ 0-003 


0-065 

0-079 

-0-014 



Sub-table b. 



Vn+l - Vn-1 

M^n+1 - I'n-l) 

v'„+i-v'n = dA' 

Vn-l + i{Vn+l - Vn-l) = v' 

dA' - v' 



10-126 



15-412 
5-137 
5-121 
5-005 

+ 0-116 



7-591 
2-530 
2-525 
2-480 
+ 0-045 



3-764 


1-871 


0-934 


0-460 


0-216 


1-255 


0-624 


0-311 


0-153 


0072 


1-252 


0-626 


0-305 


0-150 


0-068 


1-228 


0-620 


0-297 


0-147 


0-079 


+ 0-024 


+ 0-024 


+ 0-008 


+ 0-003 


-0-011 



Sub-table c. 



v"n+l-v"n = dA" 

t±j:=v" 

2 
dA" - v" 



10-305 

10-096 

+ 0-209 



5-099 

4-997 

+ 0102 



2-520 


1-251 


0-6-23 


0-309 


0-149 


0-070 


2-477 


1-2-26 


0-603 


0-294 


0-145 


0-075 


+ 0-044 


+ 0-025 


+ 0-020 


+ 0-015 


+ 0-004 


-0-005 



124 



MR J. Y. BUCHANAN ON THE 



Tables of Class F, illustrating the Method of arriving at the Volumetric Effect 
produced by changing the Concentration of a Solution. 

TRIAD OF BROMIDES. 

Table IV. POTASSIUM BROMIDE. KBr= 119-1. T = 19-5°C. 



9 

1/4 



-690 



1/8 
4-314 



7 


6 


5 


4 


3 


2 


/1 6 


1/32 


1/64 


1/128 


1/256 


1/512 


2-153 


1-081 


0-554 


0-278 


0-139 


0-074 



1 

1/1024 



Sub-table a. 



A-1000 = y 



8-857 

8-690 

+ 0-167 



4-376 

4-314 

+ 0-062 



2-161 

2-153 

+ 0-008 



1-072 

1-081 

-0-009 


0-5-27 

0-554 

-0-027 


0-276 

0-278 
-0-002 


0-139 
0-139 

0-000 


0-065 

0-074 

-0-009 



Sub-table 6. 



■y„+i - i;„-i 

v'n+i -v'n = dA' 

Vn-1 + M'""+l - •""-!) = '''' 
rfA' - v' 



8-725 



13-233 

4-411 

4-393 

4-332 

+ 0-061 



6-537 


3-233 


1-599 


0-803 


0-415 


0-204 


2-179 


1-078 


0-533 


0-268 


0-138 


0-068 


2-173 


1-072 


0-541 


0-269 


0-135 


0-068 


2-159 


1-087 


0-546 


0-277 


0-142 


0-074 


+ 0-014 


-0-015 


-0-005 


-0-008 


-0-007 


-0-006 



Sub-table c. 



v"n+\ - v"n = dA" 
v'+V_ ,1 

dA"-v" 



8-840 

8-707 

-t 6-133 



4-384 

4-323 

+ 0-061 



2-167 

2-156 
+ 0-011 



1-072 


0-534 


0-273 


0-137 


0-066 


1-084 


0-550 


0-277 


0-140 


0-074 


-0-012 


-0-016 


-0-004 


-0-003 


-0-008 



Table V. RUBIDIUM BROMIDE. RbBr = 165-5. 

Sub-table a. 



T = 19-5° C. 



■Wn+l - Vn = dA 

A-1000 = 'y 

dA-v 



10-279 

9-983 

+ 0-296 



5-042 

4-941 

+ 0-101 



2-485 

2-456 

+ 0-029 



1-234 

1-222 
+ 0-012 


0-595 

0-6-27 

- 0-032 


0-319 

0-308 

+ 0-011 


0-119 

0-189 

-0-070 


0-099 

0-090 

+ 0-009 



0-008 

0-082 

•0-074 



Sub-table b. 



'V71+I - Vii-l 

k(Vn-i-l - V„-i) 

v'n+l - v'n = dA' 

Vn-l + i{Vn+l -Vn-\) = V' 

dA' - v' 



10-048 



15 


321 


5 


107 


5 


083 


4 


965 


+ 


118 



7-527 
2-509 
2-503 
2-462 
+ 0-041 



3-719 


1-829 


0-914 


0-438 


0-218 


1-240 


0-610 


0-305 


0-146 


0-073 


1-225 


0-624 


0-278 


0-172 


0-045 


1-237 


0-613 


0-335 


0-163 


0-118 


-0-012 


+ 0-011 


-0-057 


+ 0-009 


-0-073 



0-107 
0-036 
0-036 
0-082 
-0-046 



Sub-table c. 



v"n+i -v'\^ — dA" 
v' + v 

dA"-v" 



= v 



10-247 
10-015 
+ 0-232 



5-062 

4-953 
+ 0-109 



2-494 

2-459 

+ 0-035 



1-230 


0-609 


0-299 


0-145 


0-072 


1-229 


0-620 


0-321 


0-176 


0-104 


+ 0-001 


-0-011 


-0-022 


-0-031 


-0-032 



0-022 

0-082 

-0-060 



Table VI. CESIUM BROMIDE. CsBr = 213-0. T=19-5°C. 

Sub-table a. 



Vn+l-Vn = dA 

A -1000 = 1; 
dA-v 



11-894 
11-756 
+ 0-138 



5-954 

5-802 

+ 0-152 



2-929 

2-873 

+ 0-056 



1-407 

1-466 

-0-059 


0-771 

0-695 

+ 0-076 


0-302 

0-393 

-0-091 


0-169 

0-224 

-0-055 


0-116 

0-108 

+ 0-008 



0-045 

0-063 

-0-018 



Sub-table 6. 



Vn+l - Vn-l 

^(I'n-l-l - Vn-l) 

v'n + i-l/n = dA' 

V„-l + k{l>n+l-Vn-l)- 

dA' - v' 



11-751 



17-848 

5-949 

5-917 

5-834 

+ 0-083 



8-883 
2-961 
2-923 
2-911 
+ 0-012 



4-336 


2-178 


1-073 


0-471 


0-285 


1-445 


0-726 


0-358 


0-157 


0-095 


1-490 


0-670 


0-370 


0-178 


0-086 


1-421 


0-751 


0-381 


0-203 


0-117 


+ 0-069 


-0-081 


-0-011 


-0-025 


-0-031 



0-161 
0-054 
0-054 
0-063 
-0-009 



Sub-table c. 



v"„+i-v"n = dA" 

2 

dA" - v" 



11-897 

11-753 

+ 0-144 



5-935 

5-818 
+ 0-117 



2-926 

2-892 

+ 0-034 



1-449 


0-720 


0-336 


0-174 


0-101 


1-443 


0-723 


0-387 


0-213 


0-112 


+ 0-006 


-0-003 


-0-051 


-0-039 


-0-011 



0-049 

0-063 

■0-014 



SPECIFIC GRAV^ITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 125 



Tables of Class F, illustrating the Method of arriving at the Volumetric ESect 
produced by changing the Concentration of a Solution. 

TRIAD OF IODIDES. 

Table VIL POTASSIUM IODIDE. KI = 166-1. T = 19-5°C. 



1/4 
11-281 



1/8 
5-574 



7 
1/16 
2-772 



6 


5 


4 


3 


2 


1/32 


1/64 


1/128 


1/256 


1/512 


1-395 


0-695 


0-347 


0-168 


0-089 



1 

1/1024 
0-040 



Sub-table a. 



Vn+\ - Vn = d£i. 

A-1000 = j; 

fZA - V 



11-497 

11-281 

+ 0-216 



5-707 

5-674 

+ 0-133 



2-802 

2-772 

+ 0-030 



1-377 

1-395 

-0018 


0-700 

0-695 

+ 0-005 


0-348 

0-347 

+ 0-001 


0-179 

0-168 

+ 0-011 


0-079 

0-089 

-0-010 



0-049 

0-040 

+ 0-009 



Sub-table b. 



w'„+l-l/n = C?A' 

V,i-\ + KVn+\ - Vn~l) = v' 

dA' -v' 



11-309 



17-204 

5-735 

5-701 

5-608 

+ 0-093 



8-509 


4-179 


2-077 


1-048 


0-527 


0-258 


2-836 


1-393 


0-692 


0-349 


0-176 


0-086 


2-820 


1-401 


0-691 


0-352 


0-169 


0-092 


2-788 


1-387 


0-696 


0-344 


0-175 


0-083 


+ 032 


+ 0-014 


-0-005 


+ 008 


-0-006 


+ 0-009 



0-128 
0-043 
043 
0040 



Sub-table c. 



v' + f _ 
dA" - 1 



nlA" 



11-483 

11-295 
+ 0-183 



5-704 


5-591 


+ 0-113 



2-811 

2-780 
+ 0-031 



1-389 


0-696 


0-350 


0-174 


0-085 


1-391 


0-695 


0-345 


0-171 


0-086 


-0-002 


+ 0-001 


+ 0-005 


+ 0-003 


-0-001 



0-043 
0-043 
0-000 



Table VI IL RUBIDIUM IODIDE. Rbl = 212-5. T = 19-5°C. 

Sub-table a. 



v„+i-v„=dA 
A-1000 = 'u 

dA - V 



12-969 
12-836 
+ 0-133 



6-412 

6-424 

■0-012 



3-221 

3-203 

+ 0-018 



1-601 

1-602 

-0-001 


0-789 

0-813 

-0-024 


0-391 

0-422 

-0 031 


0-204 

0-218 

-0-014 


0-075 

0-143 

-0-068 



0-082 

0-061 

+ 0-021 



Sub-table b. 



■Wn+l -Vn-1 

M^d+l - ■"n-l) ^ 

v',i+\-v'n = dA' 

Vn-l + i{Vn+l-Vn-l)-v' 

dA' - v' 



12-884 



19-381 

6-460 

6-470 

6-414 

+ 0-056 



9-633 


4-822 


2-390 


1-180 


0-595 


0-279 


0- 


3-211 


1-607 


0-797 


0-393 


0-198 


0-093 


0- 


3-205 


1-599 


0-795 


0-399 


0-180 


0-1-23 


0- 


3-209 


1-610 


0-815 


0-416 


0-236 


0-113 


0- 


-0-004 


-0-011 


-0-020 


-0-017 


-0-056 


+ 0-010 


-0- 



157 

052 
052 
061 



Sub-table c. 



v"n+i - v"n = dA" 
v'-i-V_ „ 
~2 " 

dA" - v" 



12-945 

12-860 
+ 0-085 



6-441 

6-419 
+ 0-022 



3-213 


1-600 


0-792 


0-395 


0-192 


099 


3-206 


1-606 


0-814 


0-419 


0-227 


0-128 


+ 0-007 


-0-006 


-0-022 


-0-024 


-0-035 


-0-029 



0-067 

0-061 

+ 0-006 



Table IX. CESIUM IODIDE. Csl = 260-0. T^ 

Sub-table a. 



19-5° C. 



V„+l-Vn = dA 

A-1000 = 'y 
dA -V 



14-893 
14-788 
+ 0-105 



7-445 

7-343 

+ 0-102 



3-668 

3-675 

■0-007 



1-861 

1-814 

+ 0-047 


0-875 

0-939 

-0-064 


0-455 

0-484 

-0-029 


0-207 

0-277 

-0-070 


0-042 

0-235 

-0-193 



0-082 
0-153 
0-071 



Sub-table 6. 



v„+i - V„-i 

i{Vn+l-Vn-l) 

v'n+i-v'n = dA' 

Vn-l + i(v«+l - ■»n-l) = V' 

dA' - v' 



14-789 



22-338 
7-446 
7-410 
7-379 

+ 0-031 



11-113 

3-704 
3-722 
3-657 



5-529 
1-843 
1-806 
1-851 



+ 0-065 -0-045 -0-003 



2-736 
0-912 
0-924 
0-927 



1-330 
0-443 
0-429 
0-498 
-0-069 



•662 
-221 
•180 
-318 
-138 



0-249 
0-083 
0-124 
0-194 
■0-070 



Sub-table c. 



v"„+i-v"„-dA'' 

2 

dA" - v" 



14-893 

14-788 
+ 0-105 



7-427 

7-361 

+ 0-066 



3-695 

3-666 

+ 0-029 



•124 
•041 
•041 
•153 
-112 



1-834 


0-899 


0-442 


0-194 


0-083 


1-832 


0-933 


0-491 


0-297 


0-214 


+ 0-002 


-0-034 


-0-049 


-0-103 


-0-131 



0-061 

0-153 

■0-092 



126 



MR J. Y. BUCHANAN ON THE 



§ 52. Tables of Class F, illustrating the Method of arriving at the Volumetric Effect 
produced by changing the Concentration of a Solution. 

TRIAD OF CHLORATES. 

Table X. POTASSIUM CHLORATE. KC103 = 122-6. T=19-5°C. 



9 

1/4 
11-352 



1/8 
5-632 



7 
1/16 
2-785 



6 


5 


4 


3 


2 


1/32 


1/64 


1/128 


1/256 


1/512 


1-337 


0-661 


0-324 


0-158 


0-057 



1 

1/1024 



Sdb-table a. 



A-1000 = » 
(/A - V 



11-352 



5-720 

5-632 

+ 0-088 



2-847 

2-785 

+ 0-062 



1-448 

1-337 

+ 0-111 



0-676 

0-661 

+ 0-015 



0-337 

0-324 

+ 0-013 



0-166 

0-158 

+ 0-008 



0-101 

0-057 

+ 0-044 



Sub-table h. 



J(!'„+i-1)„_,) 

v'n-\r\-v'n = dl^' 

Vn-X + \{Vn+\ - Vn-\) = v' 

dti! - v' 



11-352 



5-711 

5-641 

+ 0-070 



8-567 


4-295 


2-124 


1-013 


0-503 


0-267 


2-856 


1 -432 


0-708 


0-338 


0-168 


0-089 


2-872 


1-400 


0-707 


0-336 


0-180 


0-089 


2-769 


1-369 


0-662 


0-3-26 


0-146 


0-057 


+ 0-103 


+ 0-031 


+ 0-045 


+ 0-010 


+ 0-034 


+ 0-032 



Sub-table c. 



■y"ii+: -■y"„ = f/A" 
v'+v_ I, 

dA" - v" 



11-352 



5-716 

5-636 

+ 0-080 



2-859 

2-777 

+ 0-082 



1-424 


0-692 


0-336 


0-173 


0-095 


1-353 


0-661 


0-325 


0-152 


0-057 


+ 0-071 


+ 0-031 


+ 0-011 


+ 0-021 


+ 0-038 



Table XI. RUBIDIUM CHLORATE. 

Sub-table a. 



RbC103= 169-0. 



T = 19-5° C. 



Vn+l-Vn = dA 

A-1000 = t; 
dA-v 



12-726 



6-373 

6-353 

+ 0-020 



3-170 

3-183 

■0-013 



1-599 

1-584 
+ 0-015 



0-809 

0-775 

+ 0-034 



0-375 

0-400 

-0-025 



0-200 
0-200 
0-000 



0-089 

0-111 

■0-022 



Sub-table b. 



V„+i - V„-i 

M^»+i-""-i) 
■y',,+1 - v'„ = dA' 

dA' - v' 



12-726 



6-362 

6-364 

-0-002 



9-543 


4-769 


2-408 


1-184 


0-575 


0-289 


3-181 


1-590 


0-803 


0-395 


0-192 


0-096 


3-190 


1-596 


0-783 


0-403 


0-185 


0-096 


3-174 


1-578 


0-795 


0-392 


0-207 


0-111 


+ 0-016 


+ 0-018 


-0-012 


+ 0-011 


-0-022 


-0-015 



Sub-table c. 



v' + v 
dA" - v" 



dA" 



V 



12-726 



6-368 

6-358 

+ 0-010 



3-180 

3-178 

+ 0-002 



1-597 


0-796 


0-389 


0-193 


0-092 


1-581 


0-785 


0-396 


0-203 


0-111 


+ 0-016 


+ 0-011 


-0-007 


-0-010 


-0-019 



Table XII. CESIUM CHLORATE. CsClOg^ 

Sub-table a. 



216-5. T=19-5°C. 



Vn+\ - V„ = dA 

A-1000 = u 
dA-v 



14-515 



7-282 

7-233 

+ 0-049 



3-564 

3-669 

■0-105 



1-865 

1-804 

+ 0-061 



0-833 

0-971 

-0-138 



0-495 

0-476 

+ 0-019 



0-184 

0-292 

■0-108 



0-080 

0-212 

■0-132 



Sub-table 6. 



J(r„ + 1-D„_l) 

v'n+\-v'n = dA' 

Vn-\+h{'«n + \-'en-\)-v' 

dA' -v' 



14-515 



7-231 

7-284 
-0-053 



10-846 


5-429 


2-698 


1-328 


0-679 


0-264 


3-615 


1-810 


0-899 


0-443 


0-226 


0-088 


3-670 


1-744 


0-951 


0-401 


0-218 


0-088 


3-614 


1-870 


0-919 


0-518 


0-300 


0-212 


+ 0-056 


-0-126 


+ 0-032 


-0-117 


-0-082 


-0-124 



Sub-table c. 



v'n-yi-1>"n = dA" 
t)' + D_ ,/ 

~i " 

dA"-v" 



14-515 



7-257 
7-258 
0-001 



3-617 

3-641 

-0-024 



1-804 


0-892 


0-448 


0-201 


0-084 


1-837 


0-945 


0-497 


0-296 


0-212 


-0-033 


-0-053 


-0-049 


-0-095 


-0-128 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 127 



Tables of Class F, illustrating the Method of arriving at the Volumetric Effect 
produced by changing the Concentration of a Solution. 

TRIAD OF BROMATES. 
Tablb XIIL POTASSIUM BROMATE. KBrO, = 167-1. T = 19-5°C. 



9 

1/4 
11-290 



1/8 
5-576 



7 
1/16 
2-761 



6 
1/32 
1-370 


5 
1/64 
0-688 


4 
1/128 
0-347 


3 

1/256 
0-176 


2 
1/512 
0-089 



Sub-table a. 



1 

1/1024 



A-1000 = i; 
rfA -r 



11-290 



5-714 
5-576 

+ 0-138 



2-815 

2-761 

+ 0-054 



1-391 

1-370 

+ 0-021 



0-682 

0-688 

■0-006 



0-341 j 0-171 

0-347 1 0-176 

-0-006 1 -0-005 



0-087 

0-089 

- 0-002 



Sub-table h. 



^(v„+i - Vn-\) 
V'n+i-v'n = di^' 

clA' - v' 



11-290 



5-686 

5-604 

+ 0-082 



8-529 


4-206 


2-073 


1-023 


0-512 


0-258 


2-843 


1-402 


0-691 


0-341 


0-171 


0-086 


2-832 


1.393 


0-691 


0-341 


0-172 


0-086 


2-772 


1-379 


0-688 


0-347 


0-175 


0-089 


+ 0-060 


+ 0-014 


+ 0-003 


-0-006 


-0-003 


-0-003 



Sub-table 



v"„+l -v"n = dA" 
v' + v_ ,, 

dA" -v" 



11-290 



5-700 

5-590 

+ 0-110 



2-824 


1-392 


0-686 


0-341 


0-172 


0-086 


2-766 


1-374 


0-688 


0-347 


0-175 


0-089 


+ 0-058 


+ 0-018 


-0-002 


-0-006 


-0-003 


-0-003 



Table XIV. RUBIDIUM BROMATE. RbBr03 = 213-5. T=19-5°C. 

Sub-table a. 



Vn+l -Vn = d£i 

A-1000=:V 

dA-v 



3-057 



1-516 

1-541 

-0-025 



0-774 

0-767 

+ 0-007 



0-360 

0-407 

■0-047 



0-216 

0-191 

+ 0-025 



0-095 

0-096 

■0-001 



Sub-table b. 



M'Wn+l - Vn-i) 

v'n+i~v'n = dA' 

Vn-1 + i{Vn+l - Vn-l) = v' 

dA' - v' 



3-057 





2-290 


1-134 


0-576 


0-311 




0-763 


0-378 


0-192 


0-104 


1-527 


0-745 


0-402 


0-183 


0-104 


1-530 


0-785 


0-383 


0-200 


0-096 


-0-003 


-0-040 


+ 0-019 


-0-017 


+ 0-008 



Sub-table c. 




3-057 



1-522 


0-759 


0-381 


0-200 


0-099 


1-535 


0-776 


0-395 


0-195 


0-096 


-0-013 


-0-017 


-0-014 


+ 0-005 


+ 0-003 



Table XV. CESIUM BROMATE. CsBr03 = 261-0. T=19-5° C. 



Sub-table a. 



I'n-f-l -Vn = dA 

A -1000 = 1) 
dA -V 



3-511 



1-744 

1-767 

■0-023 



0-903 

0-864 

+ 0-039 



0-443 

0-421 

■0-022 



0-186 0-101 

0-235 0-134 

-0-049 -0-033 



Sub-table b. 



Vn+l - Vn~l 

v'n+i - v'n = dA' 
Vn-1 + i{l>n+l - Vn-i) = v' 

dA' - v' 



3-511 





2-647 


1-346 


0-629 


0-287 




0-882 


0-449 


0-210 


0-096 


1-765 


0-876 


0-425 


0-215 


0-096 


1-746 


0-870 


0-445 


0-230 


0-134 


+ 0-019 


+ 0-006 


-0-020 


-0-015 


-0-038 



Sub-table c. 



v"n+i -v"n = dA" 

2 
dA"-v" 



3-511 



1-755 


0-889 


0-434 


0-201 


0-098 


1-756 


0-867 


0-433 


0-232 


0-134 


-0-001 


+ 0-022 


+ 0-001 


-0-031 


-0-036 



128 



MR J. Y. BUCHANAN ON THE 



Tables of Class F, illustrating the Method of arriving at the Volumetric Effect 
produced by changing the Concentration of a Solution. 

TRIAD OF lODATES. 

Table XVI. POTASSIUM lODATE. KIO^ = 214-1. T = 19-5°C. 



n 
m 

V 


9 

1/4 
8-832 


8 

1/8 
4-339 


7 
1/16 
2-189 


6 

1/32 
1-096 


5 

1/64 
0-584 


4 
1/128 
0-269 


3 

1/256 
0-127 


2 

1/512 
0-057 


1 

1/1024 


Sub-table a. 


Vn+\-Vn=d£i. 

A-1000 = y 
d6.-v 


8-832 


4-493 

4-339 

-fO-154 


2-150 

2-189 

-0-039 


1-093 

1-096 

-0-003 


0-512 

0-584 

-0-072 


0-315 

0-269 

+ 0-046 


0-142 

0-127 

-f 0-015 


0-070 

0-057 

+ 0-013 




Sub-table h. 



J(%+1 -- Vn-i) 
v'n+i-v'n = dA' 
Vn-\ + k{V„+\ - Vn-\)-- 

dA' - v' 



8-832 



4-429 

4-403 

+ 0-026 



6-643 


3-243 


1-605 


0-827 


0-457 


0-212 


2-214 


1-081 


0-535 


0-276 


0-152 


0-071 


2-226 


1-058 


0-574 


0-266 


0-151 


0-071 


2-177 


1-119 


0-545 


0-279 


0-128 


0-057 


0-049 


-0-061 


+ 0-029 


-0-013 


+ 0-023 


+ 0-014 



Sub-table c. 



n-l-l -V n 
v' + V 

dA" -'o' 



dA" 



-V 



8-832 



4-461 

4-371 
+ 0-090 



2-188 


1-076 


0-543 


0-290 


0-147 


0-070 


2-183 


1-107 


0-564 


0-274 


0-127 


0-057 


+ 0-005 


-0-031 


-0-021 


+ 0-016 


+ 0-020 


+ 0-013 



Table XVII. RUBIDIUM lOD ATE. RbI03 = 260-5. 

Sub-table a. 



T = 19-5° C. 



Vn+l-1>n = dA 

A-1000 = i; 
dA-v 



2-576 



1-300 


0-615 


0-317 


0-154 


0-118 


1-276 


0-661 


0-344 


0-190 


0-072 


+ 0-024 


-0-046 


-0-0-27 


-0-036 


+ 0-046 



Sub-table h. 



Vn+l - Vn-l 
M-Wn+l - V„-i) 
v'„+i-v'n = dA' 
Vn-l + hi'^'n+l-Vn-l)-- 

dA' - v' 



2-576 





1-915 


0-932 


0-471 


0-272 




0-638 


0-311 


0-157 


0-091 


1-277 


0-644 


0-308 


0-184 


0-091 


1-299 


0-655 


0-347 


0-163 


0-072 


-0-022 


-0-011 


-0-039 


+ 0-021 


+ 0-019 



Sub-table c. 



v"„+i-v"n = dA'' 
v' +v 
~2" 
dA" - v" 



- = v 



2-576 



1-289 


0-629 


0-313 


0-169 


0-104 


1-287 


0-658 


0-345 


0-176 


0-072 


+ 0-002 


-0-029 


-0-032 


-0-007 


+ 0-032 



Table XVIII. CESIUM lODATE. CsIOj^ 308-0. T = 19-5°C. 

Sub-table a. 



Vn + l -Vn=dA 

A-1000 = v 
dA-v 



2-903 



1-432 

1-471 

-0-039 



0-685 

0-786 

•0-101 



0-329 

0-457 

-0-128 



0-185 

0-272 

•0-087 



0-120 

0-152 

•0-032 



Sub-table b. 



I'n+l - Vn-l 
v'„+i - v'n = dA' 

dA' - v' 



2-903 





2-117 


1-014 


0-514 


0-305 




0-706 


0-338 


0-171 


0-102 


1-411 


0-697 


0-352 


0-189 


0-102 


1-492 


0-795 


0-443 


0-254 


0-152 


-0-081 


-0-098 


-0-091 


-0-065 


-0-050 



Sub-table c. 



i/'n+i-i>"n = dA" 
v' + v _ ,, 

2 
dA"-v" 



2-903 



1-422 

1-481 

■0-059 



0-691 

0-790 

•0-099 



0-340 

0-450 

-0-110 



0-187 

0-263 

-0-076 



0-111 

0-152 

•0-041 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 129 



Summary. 
§ 53. Volumetric Effect produced on changing the Concentration of a Solution. 



n 
m 


9 

1/4 


8 
1/8 


7 
1/16 


6 

1/32 


5 

1/64 


4 
1/128 


3 

1/256 


2 
1/512 


1 

1/1024 


MR 


dA"-v". 


KCl 

RbCl 

CsCl 

KBr 

RbBr 

CsBr 

KI 
Rbl 

Csl 


+ 0-159 
+ 0-185 
+ 0-209 

+ 0-133 
+ 0-232 
+ 0-144 

+ 0-183 
+ 0-085 
+ 0-105 


+ 0083 
+ 0-090 
+ 0-102 

+ 0-061 
+ 0109 
+ 0-117 

+ 0-113 
+ 0-022 
+ 0066 


+ 0-037 
+ 0-026 
+ 0-044 

+ 0-011 
+ 0-035 
+ 0-034 

+ 0-031 
+ 0-007 
+ 0-029 


+ 0007 
+ 0-017 
+ 0-025 

-0-012 
+ 0001 
+ 006 

-0-002 
-0-006 
+ 0-002 


- 0-008 
+ 0020 
+ 0-0-20 

-0-016 
-0-011 
-0003 

+ 0-001 
-0022 
-0034 


+ 0-002 
+ 0-009 
+ 0-015 

- 0-004 
-0-022 
-0-051 

+ 0-005 
-0024 
-0 049 


-0-001 
-0-035 
+ 0-004 

-0-003 
-0031 
-0-039 

+ 0-003 
-0-035 
-0-103 


-0-022 

- 0-008 

- 0-005 

- 0-008 
'-0032 

-0-011 

-0001 
-0-029 
-0-131 


-0-060 
-0-014 

0-000 
+ 0006 
-0-092 


MRO3 


dA"- v". 


KCIO3 

RbClOg 

CsClOg 

KBr03 

RbHrOj 

CsBrOs 

KIO3 

RbtOg 

CsIOj 




+ 0-080 
+ 0-010 
-0-001 

+ 0-110 
+ 0-090 


+ 0-082 
+ 0002 
-0 024 

+ 0-058 
+ 0-005 


+ 0-071 
+ 0-016 
-0-033 

+ 0-018 
-0013 
- 0-001 

-0031 
+ 0-002 
-0-059 


+ 0-031 
+ 0-011 

- 0-053 

- 0-002 
-0-017 
+ 0-022 

-0021 
-0-029 
-0-099 


+ 0011 

- 0-007 
-0-049 

-0-006 
-0-014 
+ 0-001 

+ 0-016 

- 0-032 
-0-110 


+ 0-021 
-0-010 
-0-095 

-0-003 
+ 0-005 
-0 031 

+ 0-020 
-0-007 
-0-076 


+ 0-038 
-0-019 
-0-128 

-0003 
+ 0-003 
-0-036 

+ 0013 
+ 0-032 
-0-041 





§ 54. Solutions of the Salts of the Ennead MR. — The notes deal principally with 
the character of the change in the value of d^ — v with changes of the value of min 
the different solutions. 

Table I. : — KCl. — This salt is the subject of the specimen table, § 50, and it 
has been commented on in connection with that table. 

Table II. : — RbCl. — In all three sub-tables the change of sign occurs between 
m= 1/128 and w= 1/256. There is a steady decrease in the positive values from 
m=l/4 to m= 1/128 in all three cases, excepting sub-table a at m=l/64, when an 
increase over the previous value occurs. 

The solutions of this salt show a regular decrease from high positive values of 
dA — V to small negative values at low concentrations. 

Table III. :— CsCl. — The series of values of d^ — v in the three sub-tables shows 
a very regular decrease in magnitude from m = 1/4 to m = 1/256 ; change of sign occurs 
between m= 1/256 and 1/512 with a small negative value. 

Table IV. : — KBr. — There is a rapid decrease in the value of dA — v from m— 1/4 
to m=l/l6, and between 7n=l/l6 and m=l/32 there is change of sign to a small 
negative value, which persists right on to m= 1/512, and which is the characteristic 
feature of the three sub-tables. 

TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 17 



130 MR J. Y. BUCHANAN ON THE 

It would seem to indicate that in the solutions of any concentration less than 
w=l/64 the dissolution of a small quantity of salt produces the same increment of 
displacement. 

Table V. : — RbBr. — In sub-table a the value of cZA — v decreases very rapidly 

from 0*296 at m= 1/4, and changes to a negative value between m= 1/32 and m= 1/64, 

and then oscillates between alternate high negative values and low positive values. In 

sub-table c there is a regular increase in the negative values from m = 1/64 to m = 1/512. 

We have here definite evidence of expansion on dilution. 

Table VI.: — CsBr. — The character of the decline in positive values for c^A-v 
is the same as that observed in the case of KBr and RbBr, and change of sign occurs 
between 7n= 1/16 and m= 1/32. After this there is an irregular oscillation between 
high positive and high negative values in sub-table a. 

In sub-table c the positive values persist to m= 1/32, after which change of sign 
occurs, and the negative sign remains for all values down to m= 1/512. The negative 
values reach a maximum at m= 1/128 and then decline. Here again we have definite 
evidence of expansion. 

Table VII.: — KI. — A rapid decrease in positive values occurs between m=\/A 
and w = ]/16. Then, with the exception of the small negative values of 0"01 8 and O'OIO 
at ni = 1/32 and m = 1/512 respectively, there are small positive values for c?A — v. This 
holds also in sub-table c, only that the positive and negative values for values of m below 
1/82 are less. 

The final note for KBr applies with greater force in the case of KI. 

Table VIII. : — Rbl. — In sub-table a there is a sudden fall from a high positive 
value of 0"133 for m=l/4 to a negative value of 0*012 at m=l/8. and then a 
reversion to positive at 1/16 ; after this, the values are negative, except where 
m= 1/1024. 

In sub- table c there is a more progressive decrease in positive values from m = 1/4 
to m = 1/16, and between this and m = 1/32 there is change of sign. The negative values 
increase gradually to a maximum of 0'035 at 1/256 and then decrease, and we have a 
positive value of 0*006 at 1/1024. 

The evidence of expansion is conclusive. The character of these values is com- 
parable with those of CsBr. 

Table IX. : — Csl. — In sub-table a, with the exception of the sudden fall to a negative 
value of 0*007 at w= 1/16, there is a progressive decrease to m= 1/32, where change of 
sign occurs between m= 1/32 and m= 1/64, and an increase in magnitude of negative 
values to a maximum of 0*193 at w = 1/256, and then a slight fall. 

In sub-table c there is a progressive decrease in positive values to m = 1/32 ; then 
change of sign, and progressive increase in negative values to 0*131 where m= 1/512, 
a slight decrease occurring at m= 1/1024. 

The character of values for dA — t; in all three sub-tables is the same. This is the 
most marked instance where expansion occurs. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 131 

§ 55. Solutions of the Salts of the Ennead MEO^. — The reason for the small 
number of results in the cases of RbBrOg, RblOg, CsBrOg, and CsIOg is the very sparing 
solubility of these salts in water. 

Table X. : — KClOy — The values of c?A — -y in all the sub-tables are positive. 

In sub-table a they are oscillatory in character, the maximum value being O'lll at 
m= 1/32 ; the minimum of 0-008 at w= 1/256 rises to 0-044 at m= 1/512. 

In sub-table c the maximum occurs at w=l/l6, while the minimum is at 
m= 1/128. The nature of the oscillation is the same, but less pronounced. 

Table XL : — EbClO^. — There are somewhat irregular alternations between positive 
and negative values of c^ ^^ — f in the sub-table a. They are also of appreciable magnitude, 
and in sub-table c the oscillations are still apparent, although a definite change of 
sign occurs at m = 1/128, when the values of d^ — v for higher values of m are positive, 
and the lower values of m are negative. 

o 

In the table for MR = RbCl the same character is observed as here in the sub-table c ; 
see note above. 

Table XII. : — CsClO^. — There are exhibited here regular alternations between com- 
paratively high positive values and high negative values for c?A — v in sub-table a, 
except at m = 1/250 and w = 1/512, both of which are negative. 

The maximum positive value is at tn = 1/32, while the maximum negative value is 
at m= 1/64. 

Owing to the greater magnitude of the negative values than the positive ones in the 
sub-table a, we obtain a complete series of negative values of comparatively high magni- 
tude in sub-table c, and with the exception of a slight diminution in the negative value 
at m= 1/128 below that where m= 1/64, there is a progressive increase of the negative 
values from the beginning to the end, giving distinct evidence of expansion on dilution. 

Table XIII. : — KBrO^. — The character of changes in the values for d^ — v in all 
three sub-tables is the same : a rapid fall in the positive values from O'lSS at m=l/8 
to 0-021 where m-= 1/32, the negative values from m= 1/64 onwards being negligible. 

This is observed in all three tables, although in sub-tables h and c the high magni- 
tude of the positive values for d^ — v\b slightly modified. 

This feature of the steady fall in positive values to a series of negligible negative 
values is also to be seen in the series for MR = KBr, with the exception that there are 
two negative quantities of appreciable magnitude in the values for o^A — v, where 
m= 1/32 and in= 1/64 in sub-table c for KBr. 

Table XIV. : — RhBrO^. — In sub-table a there is a regular alternation between 
rather high negative values and positive values, while in sub-table c there is a regular 
transition from negative to positive values, the change of sign occurring at m= 1/256, 
the general character being that of commencing with a negative value and rising to a 
positive quantity. 

This is the only instance in these tables of the occurrence of contraction from the 
highest concentration, m= 1/32, to the lowest, m= 1/512. 



132 MR J. Y. BUCHANA.N ON THE 

Table XV. : — CsBrO^. — Here an irregular feature is observed in that there is a 
negative value of 0"023 at w=l/32, which changes to a positive value of 0'039 at 
7n= 1/64, then a diminution in a positive value at m= 1/128 to the maximum negative 
value of the series of 0*049 at m= 1/256, with a slight diminution at w= 1/512. 

In sub-table c a similar series of values for (iA — v is seen, except that they are 
more regular. 

Table XVI. : — KIO^. — In sub-table a the high positive value of 0-154 for dA-v 
at m= 1/8 is changed to a negative value of 0'039 at m= 1/16, and with a diminution 
of the negative value to 0"003 at 1/32 the maximum negative value of 0"072 is reached 
at m= 1/64 ; then a transition occurs at m = 1/128 to a positive value of 0"046. After- 
wards the positive value falls away. 

There is thus a change from a high positive value to a high negative value, and 
then reversion to moderately high positive value. 

The same feature is observed in sub-table c, but it is of a more undulatory character. 

Table XVII. :—RbIO^.— The positive value of 0-024 at m= 1/32 gives place to the 
maximum negative value of 0-046 at m= 1/64 ; then there is a diminution in negative 
values leading to a positive value of 0-046 at 1/512 in sub-table a. 

The same character is observed in sub-table c, only more regular, the maximum 
negative value of 0-032 occurring at w= 1/128. 

Table XVIII. : — Cv/Og. — ^The values for dA — v in each of the three sub-tables 
constitute the most regular of all the series. All the values are negative and reach a 
maximum value at m= 1/128 in all three sub-tables, and fall away regularly for higher 
and lower values of m. 

The character of the values for CsIOg somewhat resembles those for RblOg. 

Section IX. — Notes on the Values of v for the Bnneads MR and MROg. 

§ 56. The increment of displacement produced in 1000 grams of water at 19-5° C, 
when 1/2 gram-molecule of potassium chloride is dissolved in it, is 14*001 Gj; when a 
molecularly equivalent amount of potassium bromide is dissolved in the same quantity 
of water, the increase in the displacement is 17'547Gx; when the salt in solution is 
potassium iodide, the number is 22 '7 7 8 G^. Replacing, therefore, the chlorine by 
bromine increases the displacement by 3-546 Gx ; and if the bromine be now replaced by 
iodine, there is a further increase of 5-231 G-r in the displacement; or, replacing the 
chlorine by iodine causes an increment of displacement of 8-777 G^. Proceeding in a 
similar manner with the other salts of the ennead MR and tabulating the results, we 
obtain Table I. An inspection of the table shows us that the differences for Br — CI when 
equivalent quantities of the salts of K, Rb, and Cs are dissolved in the same quantity 
of water are of the same order of magnitude till m = 1/32. The same characteristic is 
observed between the same limits of m when bromine is replaced by iodine, but the 
differences for the same gram-molecular weight of the salt are in these three series 
greater than those observed when chlorine is replaced by bromine. In the third section 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 133 



of the table, when chlorine is replaced by iodine, the numbers expressing the differences 
of the increments of displacement for the same value of m are the sum of the differences 
for Br— CI, I— Br. 

Turning now to Table II., we have the increase in the increment of displacement 
produced when the metal in combination with the same acid is varied. Here again it 
will be found that the numbers on the same line in each section are, i7iter se, of the 
same order of magnitude till m= 1/32, after which they vary more or less irregularly. 

Table L, giving the corresponding Differences between the Values of the Increments of 

Displacement, i\ caused by the Dissolution of m gram-molecule of MBr and 

MCI; MI and MBr; MI and MCI (where M = K, Rb, or Cs) in 1000 grams of 

Waterat 19-5°C. 

ENNEAD MR. 



R = 


BROMIDE-CHLORIDE. 


IODIDE-BROMIDE. 


IODIDE— CHLORIDE. 


M = 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


m. 

1/2 


3-546 


3-625 


3-249 


5-231 


5-543 


6-031 


8-777 


9-168 


9-280 


1/4 


1-791 


1-781 


1-690 


2-591 


2-853 


3032 


4-382 


4-634 


4-722 


1/8 


0-898 


0-884 


0-813 


1-260 


1-483 


1-541 


2-158 


2-367 


2-354 


1/16 


0-469 


0-436 


0-398 


0-619 


0-747 


0-802 


1-088 


1-183 


1-200 


1/32 


0-240 


0-216 


0-241 


0-314 


0-380 


0-348 


0-554 


0-596 


0-589 


1/64 


0-131 


0138 


0-091 


0-141 


0-186 


0-244 


0-272 


0-324 


0-355 


1/128 


0-061 


0070 


0-102 


0-069 


0-114 


0091 


0-130 


0-184 


0-193 


1/256 


0-041 


0067 


0-080 


0029 


0029 


0-053 


0-070 


0-096 


0-133 


1/512 


0010 


0-017 


0029 


0-015 


0-053 


0-127 


0025 


0-070 


0-156 


1/1024 










-0-021 


0-090 









Table II., giving the corresponding Differences between the Values of the Incre- 
ments of Displacement, v, caused by the Dissolution of m gram-molecule of RbR 
and KR ; CsR and RbR ; CsR and KR (where R = C1, Br, or I) in 1000 grams 

of Waterat 19-5° C. 

ENNEAD MR. 



M = 


RUBIDIUM-POTASSIU m. 


CESIUM— RUBIDIUM. 


CESIUM— POTASSIUM. 


R = 


CI. 


Br. 


I. 


01. 


Br. 


I. 


CI. 


Br. 


I. 


711. 

1/2 


2636 


2-715 


3-027 


3-764 


3-388 


3-876 


6-400 


6-103 


6-903 


1/4 


1-303 


1-293 


1-555 


1-864 


1-773 


1-952 


3-167 


3-066 


3-507 


1/8 


0-641 


0-627 


0-850 


0-932 


0-861 


0-919 


1-573 


1-488 


1-769 


1/16 


0-336 


0-303 


0-431 


0-455 


0-417 


0-472 


0-791 


0-720 


0-903 


1/32 


0-155 


0-140 


0-207 


0-219 


0-244 


0-212 


0-374 


0-384 


0-419 


1/64 


0066 


0073 


0-118 


0-115 


0068 


0-126 


0181 


0-141 


0244 


1/128 


0021 


0-030 


0-075 


0-053 


0-085 


0-062 


0-074 


0-115 


0-137 


1/256 


0-024 


0-050 


0-050 


0-022 


0035 


0-059 


0-046 


085 


0-109 


1/512 


0009 


0-016 


0-054 


0-006 


0-018 


0092 


0-015 


0-024 


0-146 


1/1024 






0-021 




-0-019 


0-092 






0-113 



134 



MR J. Y. BUCHANAN ON THE 



§ 57. Tables III. and IV. deal with the salts of the ennead MRO3, and correspond 
in arrangement with Tables I. and II. respectively; and what has been said of the 
two previous tables holds good, in general, with these two tables, but the agreement 
of the values is not so close, nor is the number of values tabulated so great. In 
Table III. the differences of the increments are nearly all negative quantities ; in 
Table IV. they are all positive. 

Table V. gives the differences of v between the oxyhalides and the halides of the 
same metal for the two enneads, MRO3 and MR, and here also the same characteristic 
agreement is noticed between the numbers on the same line in each section of the table. 
Having thus briefly explained the contents of the tables and the chief characteristics of 
the numbers in them, we will proceed to more fully discuss the effects produced on the 
displacement of 1000 grams of water at 19 '5^ C, when the constituents of the salts 
dissolved in it are changed ; and in order to compare the results of the halides with those 
of the oxyhalides, we shall confine our attention to the numbers for m^ 1/16 and less. 



Table III., giving the corresponding Differences between the Values of the Incre- 
ments of Displacement, v, caused by the Dissolution of m gram-molecule of 
MBrOg and MCIO3 ; MIO3 and MBrOg ; MIO3 and MCIO3 (where M = K, Rb, 
or Cs) in 1000 grams of Water at 19-5° C. . 

ENNEAD MRO,. 



K03 = 


BROMATE-CHLORATE. 


lODATE— BROMATE. 


lODATE— CHLORATE. 


M = 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


m. 
1/2 
1/4 
1/8 
1/16 
1/32 
1/64 
1/128 
1/256 
1/512 


- 0-062 

- 0-056 
-0-024 
+ 0033 
+ 0-027 
+ 0-023 
+ 0-018 
+ 0-032 


-0-126 
-0-043 
-0-008 
+ 0-007 
-0 009 
-0015 


-0-158 
-0-037 
-0-107 
-0-055 
-0-057 
-0-078 


-2-458 
-1-137 
-0-572 
-0-274 
-0104 
-0-078 
-0-049 
-0 032 


-0-481 
-0-265 
-0-106 
-0063 
-0001 
-0-024 


-0-608 
-0-296 
-0-078 
+ 0-036 
+ 0037 
+ 0-028 


-2-520 
-1-193 
-0-596 
-0-241 
-0 077 
-0055 
-0031 
0000 


-0-607 
- 0-308 
-0114 
-0-056 
-0-010 
-0-039 


- 0-766 
-0-333 
-0-185 
-0-019 
-0-020 
-0-050 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 135 



Table IV., giving the corresponding Differences between the Values of the Incre- 
ments of Displacement, v, caused by the Dissolution of m gram-molecule of 
RbROg and KRO3 ; CsROg and RbROg ; CsROg and KRO3 (where ROa^ClOg, 
BrOg, or IO3) in 1000 grams of Water at 19-5° C. 



ENNEAD MRO, 



M = 


RUBIDIUM— POTASSIUM. 


CESIUM-RUBIDIUM. 


CaiSIUM- POTASSIUM. 


R03 = 


CIO3. 


BrOj. 


10,. 


CIO3. 


B.O3. 


IO3. 


CIO3. 


BrOg. 


IO3. 


m. 
1/2 
1/4 
1/8 
1/16 
1/32 
1/64 
1/128 
1/256 
1/512 


1-374 
0-720 
0-398 
0-247 
0-114 
0-076 
0042 
0-054 


0-296 
0171 
0079 
0-060 
0-015 
0-007 


0-387 
0-180 
0-077 
0-075 
0-063 
0015 


1-789 

0880 
0-486 
0-220 
0-196 
0076 
0-092 
0-101 


0-454 
0-226 
0-097 
0-014 
0-044 
0-038 


0-327 
0-195 
0-125 
0-113 
0-082 
0-U80 


3-163 
1-600 
0-884 
0-467 
0-210 
0-152 
0-134 
0-155 


0-750 
0-397 
0-176 
0-074 
0-059 
0-045 


0-714 
0-375 
0-202 
0-188 
0-145 
0095 



Table V., giving the corresponding Differences between the Values of the Incre- 
ments of Displacement, v, caused by the Dissolution of m gram-molecule of 
MCIO3 and MCI; MBrOg and MBr ; MIO3 and MI (where M = K, Rb, or Cs) 
in 1000 grams of Water at 19-5° C. 











MRO 


3-MR. 










ROs- 
R = 


CHLORATE-CHLORI DE. 


BROMATE— BROMIDE. 


lODATE- IODIDE. 


M = 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


7)1. 

1/2 




















1/4 


4-453 


4-524 


4-449 


2-600 






-2-449 






1/8 


2-216 


2-296 


2-244 


1-262 






-1-235 






1/16 


l-IOl 


1-163 


1-194 


0-608 


0-601 


0-638 


-0-583 


-0-727 


-0 772 


1/32 


0-496 


0-578 


0-579 


0-289 


0-319 


0-301 


-0-299 


- 0-326 


-0-343 


1/64 


0-238 


0-286 


0-367 


0-134 


0-140 


0169 


-0-111 


-0 152 


-0-253 


1/128 


0-107 


0-162 


0-185 


0-069 


0-099 


0028 


-0-078 


-0078 


-0-027 


1/256 


0-060 


0038 


0-148 


037 


002 


0011 


-0041 


- 0-028 


- 0-005 


1/512 


-0-007 


0-038 


0-133 


0-015 


0-006 


026 


-0032 


-0-071 


-0-083 



§ 58. If, in a solution containing I/I6 grm.-mol. KCl in 1000 grams of water at 
19*5' C, we imagine the chlorine to be replaced by bromine, the process is accompanied 
by an increase of 0'469 G-,, in the displacement of the original solution. If the operation 
be performed upon an equivalent solution of RbCl, the increase in the displacement is 
0*436 Gt; if the salt in solution be I/I6 grm.-moL CsCl, the increase is 0"398Gt- 
Therefore, it appears that, when we replace CI by Br in I/I6 grm.-mol. solutions of the 



136 MR J. Y. BUCHANAN ON THE 

chlorides of K, Rb, and Cs, approximately the same increment of displacement is 
produced. Commencing again with our solution of KCl, and replacing the potassium by 
rubidium, causes an increase in the displacement of 0-336 Gx; when rubidium is 
replaced by caesium, the increase is 0*455 Gx; replacing potassium by caesium produces 
an increase of 0791 Gx. If we now consider a solution of i/l6 grm.-mol. of KBr, and 
replace the potassium by rubidium, there is an increase in the displacement of 0'303 Gx ; 
replacing the rubidium by caesium causes a further increase of 0"4 1 7 Gx ; or if potassium 
be replaced by caesium, the increase is 0720Gt. Setting out the numbers in the 
manner shown below gives a clearer view of the various changes that take place 
when one element in a compound is replaced by another : — 

K. Diff. Rb. Diff. Cs. 

CI . . 1-684 + 0-336 = 2-o'20 + 0-455 = 2-475 
Br . . 2-153 + 0-303 = 2-456+0-417 = 2-873 



Diff. 0-469 0-436 0-398 

It will be seen that the difference between RbCl and RbBr is nearly the same as 
that between KCl and KBr, because the increase in the increment of displacement 
produced when K is replaced by Rb in KCl is about the same as that produced when 
Rb takes the place of K in KBr. Therefore, when in solutions containing 1/16 grm.- 
mol. of KCl or KBr the potassium is replaced by rubidium, approximately the same 
increase in the increment of displacement is produced ; further, when CI is replaced by 
Br in KCl and RbCl, the increase is nearly the same in each case, but of a higher value 
than when the change is made in the metals. The difference between the atomic weights 
of CI and Br is 44*5 ; between K and Rb it is 46 '4. The atomic weight of CI is less 
than that of K ; so, also, is the atomic weight of Br less than that of Rb ; yet there is a 
greater difference of displacement produced by changing the acid than by changing 
the base. 

Turning our attention now to the chlorides and bromides of rubidium and caesium, 
we find that replacing CI by Br in RbCl causes an increase in the displacement of RbCl 
which is the mean of the increases produced when Cs replaces Rb in RbCl and RbBr, 
and is greater than when Br replaces CI in CsCl. There is a greater effect produced by 
changing the metals when CI is the acid than when Br is the acid ; also, when the acid 
united with the same base is changed, the variation is greatest for the lightest metal, 
and least for the heaviest. 

Proceeding on the lines set forth above, we will next consider the changes caused by 
replacing bromine by iodine, when combined with the same three metals : — 

K. Dilf. Rb. Diff. Cs. 

Br . . 2-153 + 0-303 = 2-456 + 0-417 = 2-873 
I . . 2-772 + 0-431 = 3-203 + 0-472 = 3-675 



Diff. 0-619 0-747 0-802 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 137 



Replacing Br by I in KBr, RbBr, or CsBr produces increases in the displacements 
which become greater as the atomic weight of the metal increases. 

There is a greater increase produced by replacing K by Rb in KI than that 
produced when Rb takes the place of K in KBr ; a similar effect is seen with the 
corresponding salts of rubidium and caesium, but in both instances the replacing of one 
metal by another causes a smaller change than when iodine takes the place of bromine 
in the bromide of the metal. The exchange of iodine for bromine produces increases in 
the values of v which rise with the increase of the atomic weight of the metal in 
combination ; as was previously observed, the replacement of chlorine by bromine 
caused changes in the increase of v which diminished with the increase in the atomic 
weight of the metal. We may summarise the foregoing observations by saying that 
the replacement of the acid in a salt of the general formula MR (where M = K, Rb, 
or Cs, and R = CI, Br, or I) by another acid causes a greater change in the value of the 
displacement of a solution containing 1/16 grm.-raol. of the salt in 1000 grams of 
water at 19 '5° C. than the replacement of the metal by another metal. The only 
exceptions to this are when Cs takes the place of Rb in RbCl, and when Br replaces 
CI in CsCl, this latter being the smallest change produced when one acid is replaced by 
another. The values of the differences of the displacements for 1/16 grm.-mol. 
solutions of the salts of the halides are given in the table below. 





K. 


Diff. 


Rb. 


Diff. 


Cs. 


Diir. 
(K by Cs). 


CI 

Diff. 

Br 

Diff. 

I 
Diff CI by I 


0-469 


0-336 


0-436 


0-455 


0-398 


0-791 


0-619 


0-303 


0-747 


0-417 


0-802 


0-720 


1-088 


0-431 


1-183 


0-472 


1-200 


0-903 















In the table, differences on the same line are, inter se, comparable ; the differences in 
the same column headed " Diff." are also comparable. 

§ 59. The salts of the oxy halides will now be dealt with in a manner similar to that 
of the halides. 

K. Diff. Rb. Diff. Cs. 



C103 

BrO, 



2-785 + 0-398 = 3-183 + 0-486 = 3-669 
2-761 + 0-296 = 3-057 + 0-454 = 3-511 



Diff. - 24 



0-126 



-0-158 



If, in a solution containing 1/16 grm.-mol. KCIO3 per 1000 grams of water at 

19*5° C, we replace the chlorine by bromine, the process is accompanied by a decrease 

in the displacement of the solution by an amount equal to — 0-024 Gi ; if the operation 
TKANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 18 



138 



MR J. Y. BUCHANAN ON THE 



be performed on an equivalent solution of RbClOg, the decrease is — 0'126 Gx ; if the 
salt in solution be CsClOg, the decrease is — 0'158 Gj. Therefore, when we replace 
the chlorine by bromine in 1/16 grm.-mol. solutions of the chlorates of K, Rb, and Cs 
we observe that the change produces a decrease in the displacement of the solution, 
and this decrease becomes greater as the atomic weight of the metal increases. 

Confining our attention next to the changes produced in the displacement by- 
varying the metal combined with the same acid, we find that when K is replaced by 
Rb in KCIO3 the displacement of the solution increases, as it also does when Rb is 
replaced by Cs in RbClOg. There is a similar increase when we consider the bromates 
of the same three metals ; but replacing K by Rb in KBrOg produces a smaller increase 
than when the same change is made in KCIO3. Similarly, the replacement of Rb by 
Cs in RbBr03 is less than when Cs replaces Rb in RbClOg. In the four cases where 
an exchange of metals takes place the corresponding changes in the displacements are 
positive ; with the three changes obtained by replacing CI by Br the changes in the 
displacement are negative, and numerically less than when the metals are changed. 



K. 



Diflf. 



Rb. 



DiflF. 



BrOg 
I0„ 



2-761+0-296 
2-189 + 0-387 



Diff. - 0-572 



3-057 + 0-454 
2-576 + 0-327 

0-481 



Cs. 

V. 

= 3-511 
= 2-903 



-0-608 



With the bromates and iodates of the same three metals we find that replacing the 
bromine by iodine causes a reduction in the displacement of — 0*572 Gx in tbe case of 
KBrOg ; with RbBrOg the change in the displacement is — 0'481 G^ ; and with CsBrOg 
it is — 0'608 Gx- When we compare the results obtained by changing the metal 
combined with the same acid, we find that the iodates have positive values for the 
change in the displacement, just as the bromates and chlorates had, but that, whereas 
the replacement of Rb by Cs in RbClOg and RbBrOg gave an increase in the displace- 
ment which was greater than that produced by the exchange of Rb for K in KCIO3 and 
KBrOg, the replacement of Rb by Cs in RblOg causes an increase in the displacement 
which is less than that caused by replacing K by Rb in KlOg. The results of the 
changes produced in the displacement by the replacement of one constituent by 





K. 


Diff. 


Rb. 


Diff. 


Cs. 


Diff. 
(K by Cs). 


C103 

Diff. 

BrOs 
Diff. 

IO3 
DiffClOgbylOg 


-0-024 


0-398 


-0-126 


0-486 


-0-158 


0-884 


-0-572 


0-296 


-0-481 


0-454 


-0-608 


0-750 


-0-596 


0-387 


-0-607 


0-327 


-0-766 


0-714 















ClOj— CI . 


K. 
. 1-101 


Rb. 
1-163 


BrOj— Br . 


. 0-608 


0-601 


IO3-I . 


. -0-583 


- 0-727 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 139 

another in the salt dissolved is given in the preceding table. We do not here find so 
close an agreement between the numbers in a line, nor between those in the same 
column, but the agreement is still near enough to prevent any ambiguity as to which 
line or column a series belongs. Moreover, the columnar differences are all positive, 
while the line differences are all negative. We may further note that with the replace- 
ment of Rb by Cs and K by Cs the columnar differences decrease with an increase in 
the molecular weight of RO3, while replacing K by Rb causes irregular changes in the 
displacement. With the line differences the replacement of CIO3 by BrOg and CIO3 
by IO3 causes an increase with an increase of the atomic weight of the metal ; but with 
the replacement of BrOg by IO3 the changes in the displacement are irregular. 

§ 60. The next effect to consider is that produced by the addition of the three oxygen 
atoms to the salts of the halides to form the corresponding salts of the oxyhalides. 

Cs. 
1-194 

0-638 

-0-772 

In order to do this the above table has been constructed, in which the differences 
between the corresponding salts of the halides and oxyhalides for the same metal are 
entered in vertical columns. If we imagine that in a solution of 1/16 KCl + 1000 grams 
of water at 19-5° C. we add sufficient oxygen to the chloride and so produce KCIO3 
in solution, the operation is accompanied by an addition to the displacement of 
I'lOl Gt ; if the same operation be performed on a 1/16 grm.-mol. solution of the 
bromide of the same metal, the increase in the displacement is only 0"608 G-x ; and if 
we treat a solution of the iodide in the same way it produces a diminution in the dis- 
placement of -0"583 Gt. An inspection of the changes occurring when the three 
corresponding salts of rubidium and caesium are similarly treated shows us that they 
behave not only in an analogous manner, but that the amount of the change in each 
case is almost the same as that observed with the potassium salts, increasing slightly 
with the atomic weight of the metal. 

This action of the three atoms of oxygen upon the displacements of solutions of 
1/16 grm.-mol. of the halides in 1000 grms. of water at 19-5° C. is peculiar, since in 
each case we have added the same weight of oxygen, namely, 3 grams, and the effects 
produced by it are similar in the salts with the same acid but different bases, but differ 
when the acid in combination with the same base is varied. 

§ 61. ^ General Comparison and Summary of the Variation in the Values of 
the Mean Increment of Displacement for Dilute Solutions of Salts of the two 
Enneads MR and MRO^ {where M may he K, Rb, or Cs, and R may he CI, 
Br, or I). — This comparison includes: — 

(a) The variation produced by successive dilutions of a solution of an individual salt. 

(h) The character of the variation in the case of the whole series of solutions of 
salts of the two enneads. 

(c) The variation with the molecular weight. 



140 MR J. Y. BUCHANAN ON THE 

The first point has been adequately dealt with in the immediately preceding section, 
and the two remaining ones will now be considered. The diagram on next page 
clearly shows the relations pointed out above, and the following are the more 
distinctive features which are illustrated. 

All the halide salts, with the possible exception of KI, have the property of causing 
expansion with dilution of their respective solutions, this expansion, in the case of the 
chlorides, being nearly proportional to the rise in the atomic weights of the base, as 
shown by the almost parallel march of the curves. In the case of the bromides the 
march is not so regular, the solutions of the rubidium and caesium salts inducing a 
greater relative expansion on dilution than is the case with the potassium salt, the 
change being greatest in the case of the rubidium salt. 

With the iodides, this increased effect of expansion which occurs on dilu- 
tion of solutions of the salts of rubidium and csesium over that of potassium is 
considerably enhanced, the solutions of potassium iodide showing practically no 
expansion. 

Thus, summarising the effects, the mutual relations of halogen and base in the cases 
of halide salts of potassium and the chlorine compounds of rubidium and csesium produce 
normal effects, as shown by only slight changes in the values of vjm as the solutions 
decrease in concentration, while the remaining salts show expansion on dilution of 
solutions of them, which increases in magnitude with increase in molecular weight, 
reaching a maximum with csesium iodide. This is interesting when it is considered 
that csesium is the most electro-positive element, and seems to point to the expansive 
effect produced by both csesium and iodine independently, while mutual interference 
occurs in the other cases. 

The oxyhalides are not comparable in any sense with their respective halide 
compounds, which have been treated above. 

The most obvious feature of the incorporation of the oxygen atoms is, that the 
values for vjm decrease with the increase in molecular weight when triads of the salts 
having common base and the same concentration are compared, the only exception 
being the case of potassium bromate ; and this feature is the reverse of that in the case 
of the halide salts. Also in the case of KCIO3 and KIO3 contraction occurs on dilution 
of their respective solutions ; and where expansion occurs on dilution, the general order 
is that of proceeding from the iodates to the chlorates, where the greatest expansive 
effect is seen in the case of csesium chlorate. This is the reverse of the order which is 
seen in the case of the halides. The chlorates show the greatest variations in the 
values of vjin with dilution of solutions of the salts, and least with the bromates, the 
iodates beit)g intermediate. 

Thus the effect of the inclusion of the oxygen in the molecule of the halides is to 
greatly increase the expansion effect when solutions of the chlorates of rubidium and 
csesium are diluted, to exert very little if any effect in the case of the bromides, and to 
diminish the effect of expansion in the case of the iodides, the general effect in the 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 141 





















s 


CO 




























1 






■O 1 


o" 








































o 








































o 




















CO 




























/ 




\ 




































/ 


/ 


/ 






!i 










00 






















. 


yy 


.^ 


/ 




i^ 


\ 






























A 




\ 




/ 


\ 




\, 






"5 
















,^-- 


>^ 


, 1 




— -^ 




^--- 


-C 





\ 


*. 




















~^ 


■r-~. 










"^ 












< 


3 


C 

c 


3 
3 




3 

3) 


c 

Q 


3 



c 


3 


(1 


3 
3 


o 

in 


o 


o 


o 


o 

M 

CO 










































































1- 

J3 
K 






































9 

m 

6 


o 
m 


o 


I. 

m 










<o 




























I i 


/ 








































y 


/ 












CO 


























/ 


A 




y 




































// 


•^ \ 


\<^l 


/ 










CO 

in 




















^ 




s 




l\ 




/ 




1 




- In 




































1 




O 

CN 

CO 




o 


o 
o 




o 
oo 


O 


o S o 

lO '" "f 


o 

CO 






































































( 


1 


6 
o 
cr 






o 
o 


O 




O 


































j 






\ 




































J 








1 










03 
CN 






















^ 


^ 








\ 




; 


\ 
























^ 


^ 










\ 


J 


i 


\ 




ID 

in 
























^ 


X 






>> 


<< 


^ 




CM 51 

^in *" 











































3 

M 








































-1 


5 


o 


o 
o 


o 

0> 


o 
to 


o 


o o o 


o c 
m c 



o 



II 



S 2 



^ 

rt 



=3 

So 



142 MR J. Y. BUCHANAN ON THE 

cases of potassium and rubidium iodates being that of contraction on dilution, while 
the iodate of caesium simulates the character of the iodide of the same base, though to 
a modified degree ; as though the dominating influence were the expansive effect of the 
basic radical. 

§ 62. If we consider the solutions of the salts of the double ennead MR, MRO3, we 
have eighteen solutions for each value of m. Owing to the sparing solubility of some 
of the salts of the ennead MRO3, the highest value of m available for all the salts is 
1/16. The eighteen salts can be divided into three hexads, the members of each 
hexad containing a common metallic element, K, Rb, or Cs, and into three other hexads 
having a common metalloidal element, CI, Br, or I. The values of v ( = A— 1000) for 
the solution of 1/16 gram-molecule of each of the salts in the three hexads having 
the common elements K, Rb, Cs are arranged in the tables, and the graphic effect is 
illustrated in the diagram of § .38. When we wish to compare the solutions having 
different values of m, it is convenient to use the values of v/m, that is, the increment 
of displacement (A — 1000) reduced to the value which it would have if m = 1. This is 
found in the general tables of Class E (§ 30) ; and for the solutions of 1/32 gram-molecule 
salt and under, with nucleus CI, Br, or I, the values of vjin are represented graphically 
in the diagram § 61, in which the ordinates are values of v/m, and the abscissse 
values of m. 

When this diagram is studied, it is seen that the arrangement of the curves is 
different in each of the three compartments which correspond to the solutions of salts 
having as common elements the metalloids CI, Br, I respectively, and that their 
differences are not altogether irregular. In the first diagram, the common element 
being CI, the values of vjm follow the same order as that of the arrangement for 
m= 1/16 in the tables, namely, KCl, RbCl, CsCl, KCIO3, RbClO^, CsClOg. This order 
is maintained for m= 1/64, 1/128, 1/256. When the common element is Br or I, the 
arrangement of the salts with respect to the values of vjm is different. 

The values of vim recorded in the general tables of Class E, with the curves in 
the above diagram, furnish the means of appreciating the changing characters of the 
difierent solutions with change of concentration, having regard to the numerical values 
of the constant vjm for the different salts for the different values of m. 

§ 63. It is instructive to consider the order in which the salts of each hexad follow 
each other when arranged in ascending order of values of vjm, without paying particular 
attention to their actual numerical values. 

For this purpose it is convenient to represent each hexad of salts by a hexagon, 
the centre of which is occupied by the common element, metal or metalloid, as nucleus. 
The angles of the hexagon are then supposed to be occupied by the residues of the 
respective salts after abstraction of the common element, arranged in ascending order 
of magnitude of vjm, the lowest value occupying the lowest angle on the paper, and 
the other values of vjm occupying the other angles seriatim in ascending order of 
magnitude, and going round from left to right. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 143 



R 



In the figure we have a hexagon the corners of which are numbered on this plan 
from 1 to 6. hiside the hexagon we have the common element M 
or R, and above it the value of m for the particular solution. The 
residue corresponding to the lowest value of v/m is entered at the 
corner numbered 1, the next higher at 2, the next at 3, and so on, 
the residue corresponding to the highest value of i;/m occupying place 
No. 6. For concentrations higher than m=\/64, the arrangement of 
residues is the same as that given for m=l/64 in the six hexagons 
corresponding to the common elements CI, Br, I, K, Rb, Cs. 

In the accompanying figures we have the three hexagons 1/64 [CI], 1/64 [Br], 
1/64 [I] corresponding to the nuclei CI, Br, I, and to the value of m = 1/64. The 
salts corresponding to the first hexagon are CIM and CIMO3. and their residues after 
abstraction of CI are K, Rb, Cs, KO3, RbOg, CsOg. Entering these at the corners, on 
the plan above explained, in ascending order of magnitude of t'/m, the residue corre- 
sponding to the lowest value of v/m being entered at place 1 , we find that the order 
in which the residues follow each other is that of the salts of the two triads CIM and 



KO, 



Cs 



CsO, 



Cs 



Rb 



1/64 
CI 



RbO, 



CsO, 



K0„ 



Rb 



1/64 ^^^3 



Bp 



K 



CsO« 



K 



RbOo 



1/64 
I 



Rb 
Cs 



K 



KO, 



ClMOo, and following the ascendino; order of molecular weight in each triad. When 



the nucleus is Br, the order difi'ers from that corresponding to CI in that the neigh- 
bouring residues Cs and KO3 change places. When the nucleus is I, the arrangement 
seems to be quite diff'erent, but it is derived from that with the nucleus Br by replacing 
at each corner the residues M by MO3 and MO3 by M. 
If we arrange the residues in parallel lines we have : — 



Hexagon. 


Residues in ascending order of magnitude of v/m. 


1/64 [CI] 
1/64 [BpI 
1/G4 [I] 


K 

K 

KO3 


Rb 

Rb 

Rb03 


Cs 

KO3 

K 


KO3 

Cs 
CSO3 


RbOg 

RbOg 

Rb 


CsOg 
CSO3 

Cs 



When the metals act as nucleus, we have the hexagons in the figures on p. 144. 
In each of these hexagons CI occupies place 1, Br place 2, and IO3 place 3. The 4th 
place is occupied in the consecutive hexagons by CIO3, Br03, Br03 ; the 5th by BrOg, 
CIO3, I ; and the 6th by I, I, CIO3. The hexagon 1/64 [K] is derived from 1/64 [CI] 
by an exchange of place between I and IO3, which correspond to Cs and CsOg ; when 



144 



MR J. Y. BUCHANAN ON THE 



CIO3 and Br03 change places, we get 1/64 [Rb] ; and when CIO3 further changes places 
with I, we get 1/64 [Cs]. 



CIO, 



BrOo 



BrO, 



lOs / 1/64 \ BrO„ 



Br 



K 



IO3 ' i/g4 ^ C10„ 



Br 



Rb 



I0„ 



Br 



1/64 
Cs 



I 
CIO. 



CI 



CI CI 

The expressions 1/64 [Cs], 1/64 [CI], etc., are used as abbreviations to mean 
the hexagons corresponding to the nuclei Cs, CI, etc., and the solutions containing 
1/64 grm.-mol. salt per thousand grams of water. The general expressions 7n [R] and 
m [M] indicate the hexagons corresponding to salts with a metalloidal or a metallic 
nucleus respectively, the solutions of which contain m grm.-mol. of the salt per thousand 
grams of water. 

§ 64. It has already been pointed out that the hexagonal arrangement of residues 
in ascending order of magnitude of I'/m is the same for each nucleus at all concentrations 
for which m>l/64, and we have shown how the arrangements expressed by 1/64 [R] 
and 1/64 [M] are derived from that corresponding to 1/64 [CI]. We now proceed 
to consider the case of in [R] and m [M] when m<l/64. It is only in solutions of 
such low concentration that the phenomenon of expansion on further dilution with 
water shows itself 

Beginning with the solutions of salts having a metalloidal nucleus, R, we arrange 
the hexagons in three lines of four hexagons each. For the top lines R = CI, for the 
middle line Br, and for the lowest line I. In each line the concentrations are defined 
by m=l/64. 1/128, 1/256, and 1/512 in consecutive order. This arrangement is 
exhibited in the group of hexagons on next page. 

Considering the hexagons with nucleus CI, we see that, for m^ 1/256, the 
residues follow in the orders of the triads M, MO3, and in each triad in the ascending 
order of atomic weight, K, Eb, Cs, KO3, RbOg, CsOg. This may be called the regular 
system. For m= 1/512 this order is preserved, with the exception that place 1 is 
taken by KO3 and we have KO3, K, Rb, Cs, RbOg, CsOg. 

When the nucleus is Br and ??i=l/64, 1/128, the order of residues is that of 
consecutive triads, with transposition of KO3 and Cs. For m= 1/256 and 1/512, Rb 
and KO3 are transposed, and also RbOg and Cs, so that the residues come to be 
arranged alternately after the type M and MO3 respectively round the hexagons — 
K, KO3, Rb, RbOg, Cs, CSO3. In the eight hexagons m [CI], m [Br], place 6 is occupied 
by CSO3, and in seven out of the eight hexagons place 1 is occupied by K. 

When the nucleus is I, we find that place 1 is occupied in all cases by KO3, and 
place 6 by Cs ; that is, the initial residue of the second triad takes the lowest place, 
and the final residue of the first triad takes the highest place. Places 2 and 3 are 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 145 



occupied by RbOg and K respectively for m= 1/64, 1/128, 1/512, and by K, Rb03 for 
m= 1/256 ; places 4 and 5 are occupied by Rb and CsOg respectively for m- 1/128, 
1/256, 1/512, and by CsOg, Rb for m^ 1/64. For nucleus I and m = 1/256, the order 
of residues is the exact counterpart of that for nucleus Br and m= 1/256, 1/512. In 



Hexagons of the Type m [R]. 



KG. 



KO, 



KO, 



Cs 



Cs 
Rb 



1/64 
CI 



RbO„ 



CsO„ 



Cs 
Rb 



1/128 
CI 



K 
Cs 



RbO» 



CsO, 



Cs 
Rb 



1/256 
CI 



Rb03 
CsO, 



Rb 
K 



1/512 
CI 



K 

Cs 



K 
Rb0« 



RbO, 



CsO, 



KG, 



RbO, 



K0„ 



Rb 



1/64 
Br 



RbOg 

CsO, 



KO, 



Rb 



1/128 
Br 



RbU, 



CsO, 



Rb 
KO, 



1/256 
Br 



Cs 
CsO, 



Rb 
KO, 



1/512 
Bp 



Cs 
CsOo 



K 

CsO, 



K 
Rb 



K 
Rb 



K 

Rb 



■ K 
RbO, 



1/64 
I 



Rb 

Cs 



K 



RbO, 



1/128 
I 



KO, 



CsO, 



Cs 



RbO, 



K 



1/256 
I 



CsO, 



Cs 



K 
RbO, 



1/512 
I 



KO, 



KG, 



CsO, 



Cs 



KG, 



the latter, the residues of type MOg occupy the places with even numbers, and those 
of type M occupy those with odd numbers ; in the former, the opposite is the case, so 
that we have the following arrangement of residues in — 



Places numbered 



1/256 and 1/512 [Bp] 
1/256 [I] 



1 


2 


3 


4 


5 


6 


K 
KO3 


KO3 
K 


Rb 

RbOg 


RbOg 
Rb 


Cs 
CSO3 


CSO3 

Cs 

1 



§ 65. Turning now to solutions of salts having a metallic nucleus, M, we arrange 
the hexagons in three lines of four hexagons each, as on the following page. For the 
top line the nucleus is K, for the middle line Rb, and for the lowest line Cs. In 
each line the concentrations of the solutions are defined by to=1/64, 1/128, 1/256, 
and 1/512, in this order. 

TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 19 



146 



MR J. Y. BUCHANAN ON THE 



ClU., 



Hexagons of the Type m [M]. 

CIO, CIO, 



Br 



BrO, 




BrO, 



Br / "" T 

1/256 



Br 



^^ 1/512' ^ 



Br03 CIO, 



K 



Br0„ 



CIO, 



IO3 

BrO, 



10, 



CIO, 



10, 



Br 



1/128 
Rb 



BrOg 
1 



10, / , \ 010, Br 

^ 1 /256 ^ 



CI 



1/512 
Rb 



CIO, 



CI 
10, 



I03 

10, 



1/64 
Cs 



^^rOg ^1/128' ^^*^« ^^'0" ^-^-^ t 



CIO3 Br 



Cs 



BrOg/^^gJ^ CIO" 



ClOg Bi 



Cs 



CI 



01 



The outstanding feature of the m [M] hexagons is the position occupied by IO.3, which 
corresponds to CsOg in the m [R] hexagons. Whereas CsOg in these solutions occupies 
place 6 in eight out of twelve solutions, IO3 does not occupy place 6 in any of the m [M] 
solutions. In two cases it occupies place 1, in one case place 2, and in six cases 
place 3 ; thus, in nine cases out of twelve it is found in the first three places of the 
hexagon, and in the three remaining cases it occupies place 4. It is most frequently 
found at place 3, which in the regular system is the place for I. But I is never found 
elsewhere than in places 5 and 6. It occupies place 6 eight times, and place 5 four 
times. Therefore, the feature of the m [M] hexagons is that I and IO3 have exchanged 
their regular places. Indeed, if we change them back again in the m [K] solutions for 
which m=l/64 and 1/128, we get the regular arrangement corresponding to that of 
1/64 [CI], namely, CI, Br, I, CIO3, BrOg, IO3. This preference of IO3 for the final 
place in the first triad, in place of I, which takes that in the second triad, is quite 
comparable with the interchange of functions between K and KO3 as regards the 
initial positions in the triads of salts of nucleus R. 

The other four residues confine themselves very closely to their own triads ; thus 
CI does not leave it at all ; Br leaves it only once ; CIO3 also leaves it once, and 
BrOg three times. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 147 



§ 66. In the following tables we collect the different schemes of ordinal sequence 
of residues for the different solutions of the hexads having metalloidal nuclei, ft 
will be seen that the regukrity of these sequences and that of their progressive 
development is remarkable. 



Hexa- 
gon. 


Concentration (m) and Remarks. 


Ordinal Sequence of Residues. 


m [CI] 


The regular system of sequence is found in 1/64, 
1/128, 1/256. 


K 


Rb 


Cs 


KO3 


RbOg 


CSO3 




By shifting KO3 to the first place and closing up, we 


KO, 


K 


Rb 


Cs 


RbOg 


CSO3 




get 5/6 of the regular system. 














»i[Bp] 


When KO3 and Cs exchange places in the regular 
system, we have 1/64, 1/128. 


K 


Rb 


KO3 


Cs 


RbOg 


CSO3 




The arrangement for 1/256 and 1/512 is also a very 


K 


KO3 


Rb 


Rb03 


Cs 


CSO3 




regular one. 














mfll 


Equally regular but opposed to the last is 1/256 


KO3 


K 


RbOg 


Rb 


CSO3 


Cs 




When K and EbOg exchange places, we have 1/128 


KO, 


RbOg 


K 


Rb 


CsOg 


Cs 




and 1/512. 
















And when Rb and CsOg now exchange places, we 


KOv 


RbOg 


K 


CsO, 


Rb 


Cs 




have 1/64. 















When we consider the hexads with metallic nucleus, we find less regularity in the 
ordinal sequence of residues than we did in the hexads with metalloidal nucleus. The 
greatest regularity is shown by the hexads which exhibit the common initial sequence 
CI, Br, IO3, and we take them first. 



Hexa- 
gon. 


Concentration (m) and Remarks. 


Ordinal Sequence of Residues. 


m[K] 


1/64, 1/128 1 This arrangement is derived from the regular 
1/128 1 s?/s^em by transposition of I and IO3. 














m Rb 


01 


Br 


10, 


CIO, 


BrO, 


I 


M[Rb 


1/64, 1/256 Transposing Br03 and CIO3 .... 


CI 


Br 


10, 


BrO, 


CIO, 


I 


m Cs' 


1/64 Transposing CIO3 and I . . . . 


CI 


Br 


10, 


BrO, 


I 


CIO, 


in Cs' 


1/256 Transposing IO3 and BrOg .... 


CI 


Br 


BrO, 


10, 


I 


CIO, 


m Cs 


1/128, 1/512 Transposing CIO3 and I . . . . 


CI 


Br 


BrO, 


10, 


CIO, 


I 


/H[Rb 


1/512 Shifting IO3 to the beginning 


10, 


CI 


Br 


BrO, 


CIO, 


I 


m K 


1/512 1 These two are not derived by any simple] 
1/256 1 transpositions from others. ( 


10, 


CIO, 


CI 


Br 


I 


BrO, 


m[K] 


CI 


I03 


Br 


CIO3 


I 


BrOg 



In the following table the ordinal sequence of residues is shown in another 
form. It consists of six sub-tables corresponding to the six nuclei respectively. The 
middle column (m) gives the concentration of the solution in each line. The numbers 
in the body of each sub-table are the ordinal numbers of the places occupied by the 
residues at the top of each column in the solutions specified by the symbol of the 
hexagon at the top of the sub-table and the value of m on the same line in the column 
in the middle of the table. The numerals in the sub- tables are printed, some in black 
type and some in ordinary type. The black type is used only when the particular 
residue occurs three times out of four in its particular column ; the ordinary type is 
used when it occurs twice or once in the same column. 



148 



MR J. Y. BUCHANAN ON THE 







Residues. 






Grm.-mol. Salt 

per 1000 grms, 

Water. 






Residues. 




K. 


Rb. 


Gs. 


KO3. 


RbOj. 


CSO3. 


01. 


Br. 


I. 


CIO3. 


BrOj. 


1O3. 


Places occupied by Residues. 


m. 


Places occupied by Residues. 






Hexagon m [CI]. 












Hexagon m [K]. 




1 


2 


3 


4 


5 


6 


1/64 


1 


2 


6 


4 


5 


3 


1 


2 


3 


4 


5 


6 


1/128 


1 


2 


6 


4 


5 


3 


1 


2 


3 


4 


5 


6 


1/256 


1 


3 


5 


4 


6 


2 


2 


3 


4 


1 


5 


6 


1/512 


3 


4 


5 


2 


6 


1 






Hexagon wi [Bp]. 










Hexagon m [Rb]. 




1 


2 


4 


3 


5 


6 


1/64 


1 


2 


6 5 


4 


3 


1 


2 


4 


3 


5 


6 


1/128 


1 


2 


6 4 


5 


3 


1 


4 


2 


5 


3 


6 


1/256 


1 


2 


6 5 


4 


3 


1 


4 


2 


5 


3 


6 


1/512 


2 


3 


6 5 


4 


1 






Hexagon m [I]. 










Hexagon m [Cs]. 


3 


5 


6 


1 


2 


4 


1/64 


1 


2 


5 


6 


4 


3 


3 


4 


6 


1 


2 


5 


1/128 


1 


2 


6 


5 


3 


4 


2 


4 


6 


1 


3 


5 


1/256 


1 


2 


5 


6 


3 


4 


3 

1 


4 


6 


1 


2 


5 


1/512 


1 


2 


6 


5 


3 


4 



Section X. — Experimental Observations on the Displacement 
OF Solutions of Sodium Chloride. 

§ 67. Although chloride of sodium is not a member of either of the enneads which 
form the principal material of this research, its importance in nature justifies its 
inclusion in it. 

The observations made at 1.5'0° C. and recorded in Table No. 26, Class A, were not 
sufficient. A complete series of observations of specific gravity in which the values 
of m formed a geometric series, descending from 1 to 1/512, was made at 19*50° C. 

Besides this, two arithmetic series were included, in one of which the common 
difference in values of m was 1/128, and in the other 1/64. 

In the first of these two series there were eight solutions, in which m ^ i/128, 2/128, 
.3/128, 4/128, 5/128, 6/128, 7/128, 8/128, respectively. 

The second series proceeded by a common difference of 1/64 from 1/64 to 8/64, 
the first four of this series being included in the first arithmetic series, namely, 2/128, 
4/128, 6/128, 8/128. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 149 



Table of Results of Experiments made on Solutions of Sodium Chloride varying 
in Concentration from, 1 gram-molecule to 1/512 gram-molecule per 1000 grams 
of Water by the Hydrometric Method. 

SODIUM CHLORIDE. NaCl = 58-5. 
T=19-5°C. 



m. 



1-000 

1/2 
1/4 
1/8 

7/64 = 
6/64 = 
5/64 = 
1/16 
7/128 = 
6/128 = 
5/128= 
1/32 
3/128 = 
1/64 
1/128 
1/256 
1/512 



L 

9-1 4 y 

1 

1 

1 2-8(n!- 



1 

2 1-^3:!l 

1 ' 

2 oOWU 



4 2-6 B? 



W. 



1058-500 
1029-250 
1014-625 
1007-312 
1006-398 
1005-484 
1004-570 
1003-656 
1003-199 
1002-742 
1002-285 
1001-828 
1001-371 
1000-914 
1000-457 
1000-228 
1000-114 



S. 



log A. 



■039683 

■020283 

-010300 

■005166 

■004579 

-003920 

■003265 

-002636 

-002294 

-001947 

-0016153 

-001295|3 

-001007;3 

•000652 3 

-000325 3 

-000131 3 

-000058 3 



■0077899 

■0038002 

■0018552 

■0009262 

■0007857 

•0006761 

■0005645 

-0004411 

-00039191 

•0003444 

•00029041 

•0002312 

■0001574' 

■0001137 

0000573 ' 

■0000421 

00002431 



d log A 
dm 


A. 


dA. 


0-0079794 
0-0077800 
0-0074320 
0-0089920 
0-0070144 
0-0071424 
0-0078976 
0-0062976 
0-0060800 
0-0069120 
0-0075776 
0-0094464 
0-0055936 
0-0072192 
0-0038912 
0-0091136 


1018-099 
1008-789 
1004-281 
1002-135 
1001-811 
1001-558 
1001-301 
1001-017 
1000-903 
1000-793 
1000-669 
1000-530 
1000-364 
1000-262 
1000-132 
1000-097 
1000-056 


9-310 
4-508 
2-146 
0-324 
0-253 
0-257 
0-284 
0-114 
0-110 
0-124 
0-139 
0-166 
0-102 
0-130 
0-035 
0-041 



dA 
dm 



18-620 
18-032 
17-168 
20-736 
16-192 
16-448 
18-176 
14-592 
14-080 
15-872 
17-792 
21-248 
13-056 
16-640 
8-960 
20-992 



V 


log Aj - 3 _ 1 


m 
18-099 


log Am - 3 X 


1-000 


17-578 


2-049 


17-124 


4-199 


17-080 


8-410 


16-558 


9-914 


16-619 


11-522 


16-653 


13-800 


16-272 


17-660 


16-512 


19-877 


16-917 


22-618 


17126 


26 824 


16-960 


33-693 


15-531 


49491 


16-768 


68-512 


16-897 


135-949 


24-832 


185-033 


28-672 


320-531 



m X 

0-000 

- 0-049 
-0-199 
-0-410 
-0-771 
-0-856 
-1-000 
-1-660 
-1-591 
-1-285 

- 1-224 
-1-693 
-6 825 
-4-512 
-7-949 

+ 70-967 
+ 191-469 



§ 68. Preiparation of Solutions. — Instead of diluting a stronger solution in order 
to obtain the solution of requisite concentration in these arithmetic series, the method 
was adopted of adding a quantity of solid sodium chloride to a weighed quantity of a 
solution whose concentration was included in the series so as to produce the next 
solution of higher concentration in the series. 

The following schedule was drawn up from calculations made as to quantities of 
solution and salt required in the preparation of an arithmetic series of solutions from 
1/128 to 8/128 gram -molecule of sodium chloride in 1000 grams of water, each 
solution being prepared from the more dilute one immediately preceding it. 



a. 


h. 


c. 


d. 


e. 


/• 


n. 


III. 


Quantity of Water in 

Solution whose 

Concentration is m 

grain-molecules 

per 1000 grams Water. 


Qnaiitit}' of Salt in the 

Solution whose 

Concentration is m 

gram-molecules per 

1000 grams of Water. 


Quantity ofn-l 

Solution required, 

whose Concentration 

is m - 1 . 


Extra Quantity of Salt 

to be added to 

Quantity of Solution 

given in Column e. 


1 


1/128 


750 grams 


0-3427 grams 






2 


2/128 


730 ., 


0-6673 „ 


730-3335 grams 


0-3338 gram 


3 


3/128 


710 „ 


0-9735 „ 


710-6490 „ 


0-3245 „ 


4 


4/128 


690 „ 


1-2614 „ 


690-9461 „ 


0-3153 „ 


5 


5/1 -J 8 


670 „ 


1-5310 „ 


671-2248 „ 


0-3062 „ 


6 


6/128 


650 „ 


1-7824 „ 


651-4852 „ 


0-2972 „ 


7 


7/128 


630 „ 


2-0155 „ 


631-7275 „ 


0-2880 „ 


8 


8/128 


610 „ 


2-2303 „ 


611-9515 „ 


0-2788 „ 



Note. — The weights entered in this Table are the true weights, in vacuo, required for the production of the required solutions. 
The actual weights placed on the pan of the balance were those weights adjusted for the meteorological conditions at the 
time of experiment. 



150 



MR J. Y. BUCHANAN ON THE 



The second series of experiments was made on a series of solutions having a common 
difference of 1/64 gram-molecule. In this series it was necessary only to make experi- 
ments on solutions where ?w = 5/64, 6/64, 7/64, 8/64, as the first four solutions were 
included in the first arithmetic series. 

The remaining solutions necessary to complete the geometric series from 1 to 1/512 
gram-molecule were prepared in each case by the direct dissolution of salt in water. 

§ 69. Experimental Results. — The following table, abstracted from the full table, 
§ 67, gives the experimental results for the geometric series from 1 to 1/512 gram- 
molecule solutions : — 

NaCl=58-5. T = 19-5°C. 















d log A 


dA 


V 


m. 


W. 


s. 


A. 


dA. 


d log A. 


dm 


dm 


m 


1 


1058-500 


1-039683 


1018-099 










18-099 


1/2 


1029-250 


1-020283 


1008-789 


9-310 


0-0039897 


0-0079794 


18-620 


17-578 


1/4 


1014-625 


1-01U300 


1004-281 


4-508 


0-0019450 


0-0077800 


18-032 


17-124 1 


1/8 


1007-312 


1-005166 


1002-135 


2-146 


0-0009290 


0-0074320 


17-168 


17-080 1 


1/16 


1003-656 


1-002636 


1001-017 


1-118 


0-0004851 


0-0077616 


17-888 


16-272 


1/32 


1001-828 


1-001295 


1000-530 


0-487 


0-0002099 


00067168 


15-584 


16-960 


1/64 


1000-914 


1-000652 


1000-262 


0-268 


0-0001175 


0-0075200 


17-152 


16-768 


1/128 


1000457 


1-000325 


1000-132 


0-130 


0-0000564 


00072192 


16-640 


16-897 


1/256 


1000-228 


1000131 


1000-097 


0-035 


0-0000152 


0-0038192 


8-960 


24-832 


1/512 


1000-114 


1-000058 


1000-056 


0041 


0-0000178 


00091136 


20-992 


28-672 

1 



§ 70. Specific Gravity. — The observations were made with the two hydrometers 
3 and 17, and as a rule three series of observations were made with each hydrometer, 
so that each entry in column S is the mean of six series. The greatest departure of 
any one value from this mean was 2*2 in the fifth decimal place. 

From m = 1/2 to m = 1/8 the rate of decrease of the increment of the specific gravity 
(S- 1) is less than that of the concentration. Between m= 1/16 and m= 1/128 there 
is an approach to a proportion between the rate of decrease of specific gravity and 
that of concentration, while at ni= 1/256 and 1/512 this decrease is out of all propor- 
tion to that of concentration, 

§ 71, Displacement, A. — This value refiectsthe nature of changes in specific gravity. 
If we compare the values of v with those of d^, we see that, between m= 1/2 and 
m = 1/8, c/A is greater than v ; between m = 1/16 and m = 1/128 their values are nearly 
identical, and for m= 1/256 and m= 1/512 there is considerable disparity. 

If we take the specific gravity of NaCl in crystal to be 2-15, the sum of the separate 
displacements of 1 gram-molecule of salt and 1000 grams of water is 1027 '209. The 
displacement after dissolution has been efi'ected is 1018099, showing a contraction of 
8 '5 10, When 1/1(5 NaCl is dissolved in 1000 grams of water the increment of displace- 
ment per gram-molecule, v/m, is 16-272, showing a contraction of 10-937. Here the 
compression accompanying the act of dissolution reaches a maximum. When the 
concentration of the solution is less or greater than 1/16 Na(Jl+ 1000 grams of water the 



SPECIFIC CtRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 151 

compression accompanying the act is less in both cases. When 1/256 NaCl has been 
dissolved in 1000 grams of water the increment of displacement per gram -molecule, 
v/m, is 24"832, and when 1/512 NaCl has been dissolved in 1000 grams of water the value 
of vfm, is 28*672, which is greater than that of the salt in crystal, namely, 27 "209. 

§ 72. Experiments on Solutions forming Arithmetic Series. — Two sets of such 
experiments were made, the first series having the common difference 1/128 gram- 
molecule, while the other had a common diff"erence 1/64. 

The method of preparation of the solutions has been referred to in § 68, and in 
this connection it is only necessary to say that, owing to the interesting nature of 
the results obtained, it was considered advisable to repeat the experiments on certain 
solutions in order to check the results obtained. 

In each of these cases the solution was prepared by the direct dissolution of the 
salt in water. 

The result of these repetitions was in all cases to confirm the results obtained in 
the original experiments, and will be referred to later. 



§73. Series of Experiments on Solutions having the Common Difference dm= 1/128. 

NaCl = 58-5. T= 19-50° C. 



m. W. 


s. 


A. 


dA. 


d log A. 


d log A 

dm 


rfA 
dm' 


V 

m 


8/128 
7/128 
6/128 
5/128 
4/128 
3/128 
2/128 
1/128 


1003-656 
1003-199 
1002-742 
1002-285 
1001-828 
1001-371 
1000-914 
1000-457 


1-002636 
1-002294 
1-001947 
1001615 
1-001295 
1001007 
1-000652 
1-000325 


1001-017 
1000-903 
1000-793 
1000-669 
1000-530 
1000-364 
1000-262 
1000-132 


0-114 
0-110 
0-124 
0-139 
0-166 
0-102 
0-130 


0-0000492 
0-0000475 
0-0000540 
0-0000592 
0-0000738 
0-0000437 
0-0000564 


0-0062976 
0-0060800 
00069120 
0-0075776 
0-0094464 
0-0055936 
00072192 


14-592 

14-080 
15-872 
17-792 
21-248 
13-056 
16-640 


16-272 
16-512 
16-917 
17-126 
16-960 
15-531 
16-768 
16-897 



. § 74. Difference of Displacement, dA. — The numbers in the column dA show a 

characteristic rise in value from 7/128 to 3/128 gram-molecule concentration, and then 

a fall at 2/128 gram-molecule. 

dA 
Under ^ — these differences are referred to the constant value m=l and reproduce 

am ^ 

in an exaggerated degree the rise from the 6/128 to 3/128 gram-molecule. 

The values for v/m show a similar rise, but the maximum occurs at the value for m 

dA 
immediately preceding that for the maximum value for -^ — , but the character of the 

change is the same. 

The following table, which gives the maximum, mean, and minimum specific 
gravities for each value of m, together with the corresponding values for displacement 
and the difference of displacement for successive maximum, mean, and minimum 
displacements, confirms the reality of the character of change in the values for d/1 : — 



162 



MR J. Y. BUCHANAN ON THE 



NaCl = 58-5. T=19-5°C. 



7I>. 


Weight of Solution 
(mMR+1000). 


Maximum, Mean, and 

Minimum Specific 

Gravity. 


Displacement 

corresponding to each 

Value of S|iecific 

Gravity. 


1 
Difiference of Displace- 
ment for successive 
Maximum, Mean, and 
Minimum Values of 
Displacement. 


m. 
1/128 


W grams. 
1000-457 


S. 
1 1-000344 
- 1-000325 
1 1-000320 


A. 
1000123 
-132 
-137 


dA. 


2/128 


1000-914 


( 1-000662 
• 1-000652 
I 1-000646 


1000-252 
-262 
•268 


0-129 
-130 
-131 


3/128 


1001-371 


1 1-001022 

1-001007 

1 1-001000 


1000-349 
•364 
-371 


0-097 
•102 
•103 


4/128 


1001-828 


( 1-001305 
<' 1-001295 
( 1-001286 


1000-522 
-532 

•542 


0-173 
•168 
•171 


5/128 


1002282 


j 1-001621 
\ 1-001615 
1 1-001610 


1000-660 
•667 
•671 


0-138 
-134 
-128 


6/128 


1002-742 


( 1-001953 
<^ 1-001947 
( 1-001936 


1000^788 
•793 
-804 


0-128 
•127 
•133 


7/128 


1003-199 


1 1-002306 
J 1-002294 
( 1-002282 


1000-891 
-903 
-915 


0^103 
•110 
•111 


8/128 


1003-656 


( 1-002645 
<^ 1-002636 
(1-002623 


1001-007 
-017 
•029 


0^116 
-114 
-114 



The confirmation of the reality of the nature of the changes in displacement with 
change of concentration which is given by the preceding table is due to the close 
agreement of the specific gravity values in any series for a particular value of m, and 
the character of change in the dijEference of displacement is the same whether the 
maximum or minimum values for diff"erence of displacement A for consecutive values 
of m, or maximum value of A„^ with minimum value of ^m+i> are taken. 

§ 75. Although the above table shows that no further confirmation was required, 
the repetitions intimated in § 72 were made, and gave the following results. 

The concentrations selected for this purpose were 2/128, 3/128, and 7/128 gram- 
molecules. 

The solutions were made by the direct dissolution of the requisite amount of salt in 
the appropriate quantity of water, and the results obtained in these later experiments 
are compared in the following table, side by side with those obtained in the first series 
of experiments : — 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 153 



Concentration 
of the Solution 

in Ki-i'"- 

molecules per 

1000 Krams 

of Water. 


Results of Original Experiments. 


Results of Repeated Experiments. 


Difference 
between ihe 

Mean Specific 
Gravities in 

the two Series. 


Highest Value 
ofSiiecific 
Gravity. 


Lowest Value 
of Specific 
Gravity. 


Mean Specific 
Gravity. 


Highest Value 
of Specific 
Gravity. 


Lowest Value 
of S|iecific 
Gravity. 


Mean Specific 
Gravity. 


2/128 
3/128 

7/128 


1-000662 
1-001022 
1002306 


1-000646 
1-001000 
1-002282 


1-000652 
1001007 
1-002294 


1-000672 
1-001019 
1-002309 


1-000647 
1-000992 
1-002292 


] -000655 
1-001007 
1-002298 


0000003 
0-000000 
0-000004 



Note. — The comparison of the figures given in the above table, where the greatest difference of the mean specific gravities 
is 4 in the sixth decimal place, shows that the clianges in dis[ilacement of the solutions at tlipse concentrations are real, 
and confirm the character of the changes in the values of dA and v, as shown in the table of original results. 



%, 7Q. The Arithmetic Series of Solutions having the Common Difference of 
Concentration 1/64 gram-molecule. — Observations were made on the 5/64, 6/64, 7/64, 
and 8/64 gram -molecule concentrations, as the 1/64, 2/64, 3/64, and 4/64 gram- 
molecule concentrations were dealt with in the series having the common diiference 
1/128 {vide supra). 

The solutions were prepared according to a scheme similar to the one described 
in ^ 68. 

The following table gives all observed and derived data : — 

NaCl = 58-5. T=19-5°C. 



Weight of 
'^- : Solution. 


Specific 
Gravity. 


Displacement. 


Difference of 

Displace- 
ments. 


Diflference of 
Logarithms 
of Displace- 
ment. 


Mean Increment of 

Dis|ilaeenient per 

gram -molecule 

of Salt. 


Diiference of 
Displacement 
per gram- 
molecule. 


m. 

1/64 
2/64 
3/64 
4/64 

5/64 
6/64 
7/64 
8/64 


W. 

1000-914 
1001-828 
1002-742 
1003-656 
1004-570 
1005-484 
1006-398 
1007-312 


S. 

1000652 
1-001295 
1-001947 
1-002636 
1-003265 
1-003920 
1-004579 
1-005166 


W/S = A. 

1000-262 
1000-530 
1000-793 
1001-017 
1001-301 
1001-558 
1001-811 
1002-135 


dA. 

0-268 
0-263 
0-224 
0-284 
0-257 
0-253 
0-324 


d log A. 

-00011772 
•00011303 
-00009713 
•00012295 
-00011153 
•00010962 
•00014053 


V 

m' 
16-768 
16-960 
16-917 
16-272 
16-653 
16-619 
16-558 
17-080 


dA_ 
dm' 

17-152 
16-832 
14^336 
18^176 
16^448 
16^192 
20-736 



§ 77. Displacement (A) and Difference of Displacement {d^). — Variations in the 
values of dA. are obtained here as in the last series. 

The maximum value is 0-324 between m = 7l%^ and m = 8/64 gram-molecule 
concentration, and the minimum is 0'224 between m = 3/64 and m = 4/64, comprising, 
as it does, the interval represented in the former arithmetic series between m = 6/128 
and m = 8/128, including the two low values of 0*110 and 0*114, 

There is a sudden rise to the high value 0*284 between m = 4/64 and m=5/64, 
comparable with the change between m = 2/128 and m = 3/128 in the earlier arithmetic 
series. 

TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1> 20 



154 



ME J. Y. BUCHANAN ON THE 



The variations are more strongly marked in the values of —. — , which, in an 

am 

exaggerated manner, indicate the nature of the changes from high to low values. 

The values of v/m show an undulatory variation in consecutive values as contrasted 
dA 



with the values of 



dm' 



§ 78. In the following diagram the values of v/m are represented by the ordinates, 
and those of m by the abscissae : — 



m 












V 
















17 00 
16-50 
16 00 










/^ 


\ 












/ 






"- 








N 


\ 




/ 






/ 


















\ 


/ 




































I5S0 
15-00 
























' 


























- 





64 



3 


2 


5 


3 


7 


4 


6 


6 


7 


8 






















128 


64 


128 


64 


128 


64 


64 


64 


64 


64 



The changes in the value of v\m with change of m, as shown by the contour of the 
curve, indicate marked phases at concentrations where m = 3/128, 7/128, 4/64, and 7/64 
gram-molecules, and a very characteristic sequence of changes takes place between 
m = 4/128 and m = 8/128. 

The advantage of introducing the arithmetical series is that these remarkable varia- 
tions of displacement are clearly displayed, whereas they would be apt to be masked if 
we had only the geometric series. 

The reality of the remarkable features of solutions of low concentration ivhich have 
for the first time been demonstrated in this research is firmly established by the care 
■with which the experiments have been made, and by the agreement between the results 
of independent series of observations. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 155 

Section XL — The Principle and Construction of the Open Hydrometer. 

§ 79. AVhen the hydrometer is closed its mass cannot be diminished, and it can be 
increased only by external additions, which in practice must be immersed either in air 
or in the liquid. The use of submerged weights is attended by so much inconvenience 
that it has to be avoided ; consequently, when the instrument is closed, its mass is 
increased only by adding weights at the top of the stem. The extent to which such 
additions can be made depends on the stability of the instrument when floating in the 
experimental liquid. The instrument (No. 0) which I used exclusively during the 
voyage of the Challenger weighed, in round numbers, 160 grams, and the greatest 
weight which had to be added to it was 4*071 grams, which produced no disturbing 
effect whatever. But I had the curiosity to find out what was the limiting weight 
which could be used without altering the " trim " of the instrument, and it turned out 
that it could be used in solutions of chloride of sodium of all concentrations. Taking 
the specific gravity of the saturated solution as 1*2, 160 cubic centimetres of it must 
weigh 192 grams, so that the weight to be added was 32 grams. The instrument bore 
it ; but in a solution of chloride of calcium of slightly greater density it took a "list." 
These experiments showed that the efficiency of the hydrometer did not diminish when 
the concentration of the solution in which it was used was increased, but its handiness 
was afiectcd when such heavy weights had to be attached to the stem. 

In order to be able to use a method of such high precision for the determination of 
the density of solutions of all concentrations, I determined to construct hydrometers 
which should be left open at the top, so that their internal load, or ballast, could be 
altered, and they could then be used in exactly the same way as the closed instrument, 
by adding series of moderate weights to the top of the stem to produce corresponding 
series of immersions or displacements. 

§ 80. The glass instrument is made after the ordinary pattern, fig. 4, consisting of a 
spherical bulb at the lower extremity to hold the ballast, a cylindrical body having a 
diameter not less than that of the ballast bulb, and above it the cylindrical stem of 
relatively small calibre ; it is left open instead of being hermetically sealed as in the 
ordinary hydrometer. Instruments of this pattern may be ballasted either with mercury 
or shot. The latter is the material which has been generally used, because it is more 
easily handled than mercury, and, being confined in a spherical bulb of relatively small 
size, it cannot shift and thereby disturb the trim of the floating instrument. Moreover, 
when shot of a given and uniform size — for instance, No. 10 — is used, the load may be 
altered and adjusted by counting pellets. 

When we have to deal with concentrated solutions of salts which are at once very 
soluble and very expensive, the lower ballast bulb is suppressed, and the ballast is 
accommodated in the cylindrical body of the instrument, as in fig. 5. With this not 
uncommon form of instrument, and a cylinder of no greater diameter than that which 
is absolutely necessary to secure free flotation of the hydrometer, the specific gravity 



156 



MR J. Y. BUCHANAN ON THE 



of a solution can be determined with a minimum quantity of liquid. With instruments 
of this form it is necessary to use mercury as ballast, not only in order to keep the 






Fig. 3. 



Fig. 4. 



Fig. 5. 



centre of gravity as low as possible, but also because, when free in a wide cylinder and 
not confined in a small sphere, a quantity of shot does not necessarily rest with its upper 
surface quite perpendicular to the direction of gravity. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 157 

The paper scale, so convenient in the closed hydrometer, is inadmissible when the 
internal load has to be shifted. Consequently the scale (millimetres) is etched on the 
outside of the stem. When prepared for use the top of the stem is loosely closed by 
a thin rod of white enamel glass which passes down the stem so far as the engraved 
scale extends, and it is kept suspended in this position by being thickened to a 
button at the top. This addition was made in order to provide an opaque white 
surface behind the scale. I was unable at the time to procure suitable tubing with 
white slip let into it. It is an integral part of the instrument, to the mass of which it 
contributes its share. 

The mass of the hydrometer is equal to the sum of the masses of its parts, namely, 
the glass, the ballast, and the air respectively. 

§ 81. In order fully to appreciate the importance of each of these masses in furnishing 
the effective weight of the instrument, let us imagine that we are actually working in a 
vacuum, and at sea-level in latitude 45° ; thus the air both outside and inside falls 
away, and we have only the glass and the ballast remaining. Let us assume that 
the load is of shot and has been so adjusted that the instrument, when immersed in 
distilled water, of the temperature which is to be maintained uniform during the 
experiment, floats with only a small portion of the stem immersed. 

The first operation performed in the vacuum is to weigh the instrument ; let its 
weight be W grams (true). Experiment No. 1. — Let it then be floated in distilled 
water of the fixed constant temperature, and let the surface of the water cut the stem in 
a line at C, fig. 6, next page. Then the weight of the water displaced by the instrument 
below the line C is W grams (true). If we replace the water in the cylinder by a liquid 
of greater density, such as a saline solution, and float the instrument in it, we find 
that less of the instrument is immersed, and we are obliged to add weights at the top 
of the stem in order to immerse it until the surface of the liquid cuts the stem in 
the line C. Let the weight so added be w grams (true). The weight of the liquid dis- 
placed by the hydrometer when immersed in it up to the line C is then (W 4- w) grams 
(true). From this it follows that the weights of equal volumes of distilled water and 
of the experimental liquid, at the particular fixed temperature, are in the proportion of 

W : W -h tf, and the specific gravity of the liquid is — ^ — , referred to that of distilled 

water of the same temperature as unity. 

Experiment No. 2. — Let us now replace the experimental liquid in the cylinder 
by distilled water and float the hydrometer in it. As before, the surface of the water 
will cut the stem at C. Let us add a small weight to the stem, so that the instru- 
ment is depressed until the surface of the water cuts the stem at D. Let the weight 
of this small weight be u^ : then the weight of water so displaced up to D is (W-|-Mj) 
grams (true) ; let us now replace the water in the cylinder by the same experi- 
mental liquid as before ; and let small weights be added to the top of the stem 
until the surface of the liquid cuts it at D. Let the weight of these small weights 



158 



MR J. Y. BUCHANAN ON THE 



be iv-^. Then the weight of the liquid displaced by the hydrometer is (W + Wj) 
grams (true), and the specific gravity of the liquid at the fixed temperature is 

W + i^i 



Q 








Fig. 6. 



, . We may repeat this operation, with different added weights, 

VV *T" t^i 

as often as we please, and it is evident that at each operation we obtain 
a perfectly independent determination of the specific gravity of the 
liquid referred to that of distilled water of the same temperature as 
unity. 

Experiment No. 3. — While continuing to work in the vacuum, let 
us allow air to enter and fill the hydrometer, after which the top of 
the stem is closed air-tight by a cover without weight. We now weigh 
the hydrometer, and find its weight to be (W + a) grams (true). 

Let it be immersed in distilled water of the fixed temperature. 
Being heavier than before, by a, the weight of air which it contains, 
it will sink in the water until its surface cuts the stem in a line a 
little above C, say C ; therefore (W + a) grams (true) is the weight of 
the water displaced by the part of the hydrometer below C Let the 
water now be replaced by the same experimental liquid as before, and 
let the hydrometer be immersed in it, and let weights be added to the 
top until the surface of the liquid cuts the stem at C ; let this added 
weight be tv' grams (true), w' will be a little greater than was w in 
the first experiment by the difi'erence between the weights of the small 
cylinder CC of water and of liquid respectively, and the specific gravity 



of the liquid will be given by the ratio 



W + a + w' 
W + a ' 



which must be 



equal to — ^ — , as in the first experiment. 

Experiment No. 4. — Similarly, if the hydrometer be now immersed 
in the distilled water with weight u^, then the surface of the water 
will cut the stem at D' ; if the water is now replaced in the cylinder 
by the experimental liquid, then we shall have to add a weight iv^', a 
little greater than w^ in experiment 2, and the specific gravity of the 

liquid will be given by .^^ -, and it will be the same as in the 

^ 6 / W + a + Wi 

former experiments. 

Let us now return to the experimental conditions in which the 

hydrometer full of air floats in vacuo at C in water, and, with the added weight 

w', Sit C in the liquid. 

Experiment No. 5. — Let air be admitted generally, and let it be of the same 

density as that in the hydrometer, so that we are experimenting in air instead of in 

a vacuum. The hydrometer is still closed by the cover without weight ; and, as 



SPECIFIC GEAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 159 

the density of the air within and without the instrument is the same, there is no 
cause for disturbance of equilibrium between the two masses of air, and they will 
not interfere with each other. 

The external air, admitted generally, reaches only to the surface of the water or 
liquid and cannot interfere with the immersed portion of the hydrometer. The stem, 
however, is now surrounded by a medium of given density, whereas, before, it was 
surrounded by one of insensible density. The exposed stem displaces its own volume 
of the air, and the downward vertical pressure which it exerts on the immersed portion 
of the hydrometer is diminished by the weight of this volume of air. The whole of 
its vertical pressure is exerted on the immersed part of the hydrometer below the 
line C. As this pressure is diminished by the weight of the air displaced by the stem, 
the hydrometer will rise and will float a little higher ; let it cut the stem at the line C", 
which is situated a little lower than C but higher than C. 

We have then the proportion : — 

The volume of the cylinder : that of the cylinder : : the volume of the exposed : the volume of the air in 
C'C" C'C stem the hydrometer. 

The total vertical pressure exerted by the hydrometer when floating in the distilled 
water is now W + a — s, and the stem cuts the water at C". s is the weight of the air 
displaced by the exposed portion of the stem. 

Let the distilled water be now replaced by the same experimental liquid as before, 
and let the hydrometer be immersed in it, and let weights be added to the top of the 
stem until the surface of the liquid cuts the stem at C" ; let this added weight be w" 
grams (true), w" will be a little less than was w' in Experiment No. 3, and the 
specific gravity of the liquid is given by the ratio 

W + a + w" — s 
W + a-s ' 
which must be equal to 

W + a + w' 
W + a ' 

as in the third experiment, and also equal to — = — , as in the first experiment. 

Experiment No. 6. — Similarly, if the hydrometer be now immersed in the distilled 
water with weight u^, then the surface of the water will cut the stem at D", a little 
lower than D', but higher than D. If the water in the cylinder is now replaced by the 
same experimental liquid, then we shall have to add a weight w\, a little less than 
w\ in Experiment No. 4, and the specific gravity of the liquid will be given by the ratio 

W + a + io'\ - s 
W + a + Wj - s ' 

and it will be the same as in the former experiments. 

§ 82. The above suggested experiments will be best understood by reference to a 
specific instance of the use of the instrument. 



160 MR J. Y. BUCHANAN ON THE 

The selected example is that of hydrometer A when immersed in distilled water, 
and afterwards in a 4*4 gram-molecule solution of potassium chloride. 

Referring to Experiment No. 1, when working completely in vacuo, with the hydro- 
meter immersed in distilled water, the adjustment would be such as to cause it to Hoat 
with the stem immersed to a scale reading of C = 177 mm. 

Weight of glass and shot = W= 136-69884 grams. 

On replacing the distilled water by the experimental solution, the added weight 
necessary to cause the instrument to float at 17"7 mm. in vacuo would he w^ 23'25860 
grams (true). 

Weight of shot + glass + added weight = W + w= 136-69884-1- 23-25860= 159-95744 
grams. The specific gravity would therefore be : — 

W + w 159-95744 



W 136-69884 



= 1-170144. 



On admitting air into the hydrometer (Experiment No. 3), the result would be to 
cause the instrument to be immersed to C'= 32-2 mm. in distilled water. 

This is arrived at in the following manner : — The internal volume occupied by the 
air is 112-493 c.c, and the density of the air was 0'001208 gram per cubic centimetre. 
The weight would therefore be a = 0"13592 gram. 

Since O'l gram added weight produces an immersion of 10-69 mm. of stem, and as 
the weight of air admitted is distinctly an added weight, the immersion produced by 
0-13592 gram would be 

10-69 X 0-13592 ,,r 

= 14-0 ram. 

0-1 

Whence 17-7 H- 14*5 = 32-2 mm. 

Total weight after admission of air to hydrometer is : — 

Weight of shot -f- glass -1- air = W -t- a = 136-69884 + 0-13592 = 136-83476 grams. 

On immersing the hydrometer filled with air into the experimental liquid with a 
similar adjustment as in the first experiment {W + tv), the hydrometer would be immersed 
to a point short of C, since the air represents the same added weight in this case as 
when the hydrometer is immersed in distilled water, and the same added weight would 
not produce so great an immersion of the stem in the experimental liquid as in the 
distilled water of lower density. Hence an addition must be made to the original added 
weight (w) to cause the hydrometer to float at C, the 32-2 millimetres division, when 
immersed in the experimental liquid, and the value of this addition is the diff"erence 
between the weights of the same volume of experimental liquid and distilled water 
represented by the volume of the portion of stem immersed when air was admitted into 
the hydrometer while experimenting in distilled water. 

Then we have seen that the weight required to increase the displacement in 
distilled water from C to C is the weight of the air filling the hydrometer, namely, 
0-13592 gram. When the distilled water is displaced by the experimental solution, 



SPECIFIC GKAVITY AND DISPLACEMENT OF S01V1E SALINE SOLUTIONS. 161 

of specific gravity 1 "170144, then the air admitted depresses the hydrometer in the 
solution from C to a point lower than C The total weight required to increase the 
immersion of the hydrometer in the solution from C to C is 013592 x ri70144 
= 0'15905 gram. Therefore, in addition to the weight of the air, we require a 
supplementary weight = 0-15905- 0-13592 = 0-02313 gram. The total weight of the 
hydrometer is : — 

Glass + shot =:W= 136-69884 grams. 

Air =a = 0*13592 „ 

Total added weight = w' = 23-28173 „ 

160-11649 „ =W + a + w'. 

In this case 23-25860 + 0-02313 = 23-28173 grams = the total weight added to the 
top of the stem. 

The specific gravity under these conditions is : — 

W + a + w>' _ 160-11649 _^.^^^^^^ 
W + a "136-83476" 

When air is admitted generally (Experiment No. 5), the line of flotation will be at 
another point, C", for distilled water, and, as explained above, may be represented as 
a deduction from the added weight, the amount being equal to the weight of air 
displaced by the non -immersed portion of the stem ; let s denote the value of the 
weight of air displaced. 

The volume of the non-immersed portion of stem is 0-9 c.c , and the weight of 

1 c.c. air = 0-001208 gram. Therefore the weight of air displaced is 5 = 0-00109 gram. 

Now an added weight of O'l gram produced an immersion of 10-69 millimetres, when 

the hydrometer was immersed in distilled water, so an alteration of immersion of the 

stem will occur, the amount being 

10-69x0-00109 rt.,, 
— = 11 mm. 

Hence the final position of the hydrometer when immersed in distilled water will be 
C" = 32-09 mm. ; and the total weight : — 

Glass + shot + air = W + a = 136-69884 + 0-1 3592 = 136-83476 grams. 
Less correction for non-immersed portion of stem = s=— 0-00109 ,, 

W + a-s=136-83367 

In the case of the experimental liquid, the final position is nearly that of C", the 
actual correction for the non-immersed portion of the stem being arrived at in the same 
manner as above, since 0-1 gram added weight produced an immersion of the stem of 
9 '17 millimetres, and the volume of air displaced being the same as above, as also is the 
weight, the alteration of immersion will be 

9-17x0-00109 ^,^ 
Ty. =0'10 mm. 

TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 21 



162 MR J. Y. BUCHANAN" ON THE 

The scale reading would therefore be 32*10 millimetres, and as the two readings 
are such that the difference on the millimetre scale is imperceptible, they are taken 
as identical. 

The final weight is therefore : — 

Shot + glass + air + added weight 

= W + a + M;'=136'69884 + 0-13592 + 23-28170 = 160-11649 grams. 
Correction for non-immersed portion of stem ^s = — 0'00109 



&' 



W + a + it;'-s=:160-11540 



o -n -^ 160-11540 T.T^,ni.. 

Specific gravity = ^^^^^^^ = 1 1701 44. 



C" is then the final position at which the surfaces of the water and of the 
experimental liquid cut the stem when the added weight is nothing for water and 
w' for the experimental liquid, and the experiment is made in air. The effective 
downward vertical pressures are represented by the true weights W + a — s and 
W + a + w' -s respectively. 

It should be noted that in the example here given the value of the weight added 
to the stem of the hydrometer to cause it to float at C" = 32 09 mm. is so very nearly 
the same as that which caused it to float at C'=32-2 mm., that no alteration has 
been made in the value of this weight, and we have, therefore, used the symbol tv' in 
this first experiment instead of w", as given in Experiment No. 5. 

We have imagined that the open hydrometer was actually weighed in a vacuum, 
when it contained no air. In practice the hydrometer is weighed full of air and in air. 
When to this weight we apply the vacuum correction, that is, the weight of air displaced 
by the whole hydrometer and closed with its weightless cover, we obtain the value of 
W + a which is the working weight in vacuo of the hydrometer. In this expression, 
for any particular load, W is constant, it is the sum of the weights of glass and shot 
alone. The weight of air, a, contained in it will vary with the density of the atmosphere 
at the time. 

§ 83. The open hydrometer consists of G^ grams (true) of glass and L,, grams (true) 
of shot and A,, grams (true) of air, as when weighed in vacuo. In order to obtain these 
constants, we first weigh the glass instrument empty as it comes from the glass-blower, 
and find that it weighs G grams in air of given density. Taking the specific gravity 

G 

of the glass to be 2-5, we obtain -^ as the volume (in cubic centimetres) of the glass. 

z 

C 
The weight of -^ cubic centimetres of air of the given density is a^ grams, and when 

added to G gives the weight in vacuo of the glass of the instrument : 

Go = G + Ug, 

Similarly, the weight of the sliot added as load is found to be L grams in air, and 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 163 

if we take its specific gravity to be ir35, the volume of air which it displaces is 
c.c, the weight of which at the observed density is a^ grams, whence the weight 

1 J. oo 

in vacuo of the lead is 

L„ = L + «(. 

If the load L has been so adjusted that at the temperature, T, fixed for the experiments 

the hydrometer floats in distilled water at the top of the stem, then the weight of 

distilled water displaced by the hydrometer when so floating is in vacuo G^ + L^ + A^, 

whence the external volume of the whole instrument is obtained ; let this volume be V, 

then we have 

r^ T A 

where (p is the density of the air expressed in grams per cubic centimetre if A,, is 
expressed in grams. 

In this equation V, G^,, and L,, are known, therefore 

</.- V2-5 + 11-35J 
whence 

In any locality (p varies with the weather, but it can always be ascertained by the 

observation of the meteorological elements. The relative humidity of the air in the 

special room required for this work seldom differs much from 50 per cent. ; therefore 

the variations in the density of the air are due almost wholly to variations of the 

barometric pressure. In London, the extreme range of barometric pressure may be 

taken to be between 730 and 770 millimetres, having therefore an amplitude of 40 

millimetres. If we suppose that the barometric pressure was 750 millimetres when 

the instrument was weighed, and that the air then weighed 1 "2 milligram per cubic 

2 
centimetre, the extreme variations of density to be expected will be =t— x 1*2 = 0'032 

milligram per cubic centimetre. In the case of hydrometer No. 1, when loaded for 
work in distilled water the volume of air in the instrument was 112"5 c.c, which at 
1'2 milligram per c.c. would weigh 135 milligrams, and the extreme variations of this 
weight would be dz4"32 milligrams. 

It is evident, therefore, that the actual weight of the air in the hydrometer at the 
time of making an experiment is an essential factor in computing the weight of liquid 
which it displaces. There is therefore an advantage in making such observations when 
the weather is settled ; the variations of the barometric pressure in the course of a day 
are then of such an order as to be almost negligible. If, however, a cyclonic depression 
is passing over the locality, the change of barometric pressure from hour to hour may 
have to be taken into account. 

The dimensions of this instrument, fig. 4, are : — from lower extremity A to 



164 MR J. Y. BUCHANAN ON THE 

contraction B between the ballast bulb and the body of the instrument, 5 centimetres ; 
the body from B to C, 13 centimetres; the stem, from C to F, 14 centimetres, making 
the total length over all 32 centimetres. On the stem a length of 10 centimetres, DE, 
is divided in millimetres, and numbered at each centimetre, 0, 1, 2, ... 10, from 
below upwards; the lowest division, 0, is 1 centimetre from the junction of the stem 
with the body of the instrument at C, and the highest division, 10, is 3 centimetres 
from the upper extremity of the instrument at F. The external diameter of the body 
of the instrument is 37 millimetres, and that of the ballast bulb 32 millimetres. The 
external diameter of the stem is 3'5 millimetres, and the internal diameter 2"5 milli- 
metres. As the ballast used is lead shot, and the load of this shot has to be frequently 
altered, the internal diameter of the contraction at B as well as that of the stem must 
be such that shot can be added to or removed from the instrument without trouble. 
G is the button on the cane of white enamel glass which is suspended in the axis of 
the stem, and, by affording a white background, enables the scale which is etched on 
the glass to be seen with facility. 

The glass shell of the hydrometer, as it came from the glass-blower, was first 
weighed approximately, and the weight so found was 40 "5 grams. It was then loaded 
with No. 10 lead shot so that it floated in distilled water of 19'5° C. with the zero division, 
which is the lowest on the scale etched on the stem, exposed above water. 95 '6 grams 
of shot were required for this purpose. 

The hydrometer so loaded was found to weigh exactly 136 "1022 grams in air. 

The atmospheric conditions were as follows : — 

Barometer, 760 "3 mm. 
Temperature of air, 19"0° C. 
Relative humidity, 70 per cent. 

Whence the weight of 1 c.c. air = 0*001208 gram. 

Taking the specific gravity of glass to be 2"5, that of lead to be 1 r35, we have :— 

For the volume of 40 '5 grams glass . . . . . 16*2 c.c. 



For that of 95*6 grams lead . . . . 

And for the total volume of lead and glass 
The volume of 136"1 grams brass weights is 

Whence the balance of volume for correction is 



8-4 

24-6 
17-0 

7-6 



The weight of 7 '6 c.c. air under the above conditions is 0*00912 gram. 

Therefore the true weight in vacuo of the glass and shot is 136*1022 4- 0*00912 = 
136*11132 grams. To this has to be added the weight of the air which the hydrometer 
contains This is arrived at by floating the instrument in distilled water and by adding 
suitable weights at the top of the stem, so as to immerse it up to the 50-mm. division 
on the scale. The weight of distilled water so displaced is equal to the sum of the 
weights of the glass, the shot, the small external weights added, and that of the air 



SPECIFIC GRAVITY AND DISPLACKMENT OF SOME SALINE SOLUTIONS. 165 

enclosed. As mean of the experiments made at 19'5° C, the added external weight 
required was 075471 gram; so that the total solid weight of the hydrometer was 136"86603 
grams, which may be taken as the weight of distilled water at the temperature 19*5° 
displaced by the hydrometer. Dividing this weight by 0*99834, the density of distilled 
water at 19"5°, we obtain 1 37*093 c.c. as the volume of the water, which is equal to the 
external volume of the instrument which displaces it. If from the volume so found 
we deduct the volume of the glass and lead, we find the volume of the air contained 
in the instrument to be 112*493 c.c. Under the atmospheric conditions prevailing at 
the date of the experiment, 1 c.c. of air weighed nearly 1*2 milligram, whence we 
obtain 0*13592 gram for the weight of the enclosed air at the time. The weight of 
enclosed air is not constant. It is subject to slight variations, principally those of 
the barometric pressure, and these have to be taken into account. 

§ 84. There remains now only one item to complete the total effective weight of the 
floating hydrometer, namely, that of the air displaced by the portion of the stem above 
water when the instrument is in equilibrium with the water. The total effective weight 
of the hydrometer is diminished by this amount. When it is immersed up to the 
oO-mm. division, the portion of the instrument not immersed in water is the part of the 
stem above the 50-mm. division, having a length of 75 millimetres, namely 50 mm. to 
the upper end of the scale and 25 mm. to the top of the stem. 

The experiments made in distilled water at 19*5° C. with this hydrometer showed 
that the addition of 0*1 gram to the weight at the top of the stem increased the im- 
mersion by 10*69 mm., whence we obtain 0*7 c.c. as the volume of the exposed stem ; 
and the weight of this volume of air is found as above to be 0*00084 gram. This has to 
be deducted from the sum of the weights of glass, lead, and air. We have then : — 

Weight in vacuo of glass and lead 

,, ,, air enclosed ... 

„ ,, added external weight . 

Giving as total weight .137*00195 

Deducting weight of air displaced by stem .... -0*00084 

We have for the final weight of the loaded hydrometer floating 

at 50 mm. in distilled water at 19*5° C. . . . 137*00111 



136*11132 grams. 
0*13592 „ 
0*75471 „ 



Deducting from this the added external weight, 075471 gram, 

we obtain the true weight of the hydrometer alone . 136*24640 

Section XII. — Numerical Details illustrating the Use of the Open Hydrometer. 

§ 85. For this purpose we will consider in detail the items of the experimental 
determination of the weight of liquid displaced by the instrument which is designated 
" Hydrometer A " ; it has for some time been in constant use. 

A scheme for recording the items of observation is given in § 86. 



166 MK J. Y. BUCHANAN ON THE 

In the first vertical columii the line corresponding to each item is designated by 
a letter — a, h, etc. In the second column are the symbols used for the principal items, 
and in the body of the scheme the items are described or explained. 

At the top of the table the standard temperature, T, selected for the experiment is 
given, along with the designation of the liquid experimented on. 

The descriptions in the scheme explain all the items, but we may refer more 
particularly to one or two of them. 

In lines d and k we have the times of the beginning and end of the experiment. 
These are important, not only with a view to ascertaining the duration of the experi- 
ment, but also as a matter of routine in all laboratory work. It is often of great 
importance in the discussion of the results of experiments to know if errors which 
appear to be possible, or indeed probable, could in fact have occurred in the time or in 
the order in which the experiments were made. 

Lines e and^, the initial and final temperatures of the liquid. As above indicated, 
these should be identical, and the condition Ti = Ty = l'' should hold. The thermometer 
chiefly used in these experiments was one graduated on the stem into tenths of a 
Centigrade degree, the length of the whole degree being 12 millimetres. This is a very 
suitable type of thermometer for the work. 

Lines y* (f^, f=^, etc.). Each of these lines contains two entries in each series, namely, 
%v^ the " added weight " in grams, and E, the corresponding division, in millimetres, on 
the stem, at which the hydrometer, when so loaded, floats in the experimental liquid. 
The addition of external weights proceeds usually by increments of 0"1 gram, and when 
distilled water was the liquid, and hydrometer A was being used, each such increment of 
weight produced an average increment of immersion equal to 10'69 millimetres, so that 
nine observations could be made in each series. In concentrated solutions as many as 
eleven single observations could be made in one series. The initial added weight was 
regulated so that the fifth or middle reading should approximate closely to the 50-mm. 
division, which was the arbitrarily selected division on the stem for the average 
immersion of the hydrometer in every series. 

By dividing the difference between the first and last of n readings by n — 1 we 
obtain the mean immersion produced by O'l gram, which is given in the line g, and we 
are thus able to determine the displacement of each millimetre of the stem. 

Lines n and o. Line n contains the correction, dw„ to be applied to the mean 
added weight w in order to make R = 50 mm. ; and line o gives w + dw,., the total 
added weight when the hydrometer floats at 50 mm., the temperature being T. 

Lines p and q. These lines contain the correction, dLVt, to be applied to the 
mean added weight w, to compensate for the difference dt of T from the standard 
temperature T. 

Line r. This line contains the sum w + div,. + diVt, which is the total added weight 
which would immerse the hydrometer to 50 mm. when floating in the liquid having 
the temperature T exactly. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 167 



In order to arrive at the value of the correction for temperature difference in terms 
of weight, two series of observations are made in the experimental solution at two 
temperatures one degree higher and lower respectively than the standard temperature 
selected for the specific gravity determination. The same added weights are used 
in both cases. 

Under these circumstances the scale readings for the same added weight in the first 
case must be higher than those in the second ; a series of differences of scale readings for 
the same added weight is obtained, which in the case of hydrometer A amounted to 5 "0 mm. 
for each pair of observations ; hence, for a difference of O'l"^ C. in the mean temperature 
indicated in line I, we arrive at a value of 0"25 m.m. in scale reading, which represents 
the effect of the alteration of the temperature of the liquid by 0'1° C, and, as has been 
indicated above, this scale reading can be interpreted in terms of added weight diVt, to 
be added to or subtracted from the total added weight according as the mean temperature, 
T, is higher or lower than the selected standard temperature, T. 

We thus obtain the total weight of experimental liquid displaced by the hydrometer 
when floating in it at the 50-mm. division. 

When the experimental liquid is distilled water, the entries in lines t and v are 
identical, and the corresponding entry in line w is unity. Before proceeding with the 
determination of the specific gravity of solutions, a number of series of observations 
are made with the instrument in distilled water at the selected standard temperature, T, 
by which we arrive at the total weight of the instrument when floating and immersed 
in this liquid up to 50 mm. on the stem. When we are using the closed hydro- 
meter this number is a constant. When the experiments in distilled water are being 
made with the open hydrometer, the weight of air actually present in it at each experi- 
ment is ascertained and taken into account in the computation of the whole displacing 
weight of the hydrometer in the given conditions. But whether the hydrometer is 
open or closed, the mass of air in it only contributes so much to the total weight of 
the instrument at the moment. It makes no difference whether the air enclosed in 
it forms a greater or less proportion of it. 

§ 86. Scheme for Logging the Observations made with the Hydrometer in the 
JExperimental Liquid at the Selected Standard Tempei-ature, T. 



Line. 
a 


Symbol. 


Explanation. 




Designation of the hydrometer used. 




b 


W 


Weight in vacuo of the closed hydrometer ; or of the glass and shot in the case 


of the open hydrometer. 


c 




Date when the experiment was made. 




d 




Time of commencement of experiment. 




e 


t 


Initial temperature of liquid. 




/o 




Added weight w in grams :— Scale reading R in millimetres. 




/i 




Wi Ri 




h 




1^2 1^2 




f} 




W>3 Rg 




h 




^4 R4 




h 




W5 Rs 




f} 




Wg Re 




f) 




w, R, 




U 




Wg Rg 




h 




Wg Rg 





168 



MR J. Y. BUCHANAN ON THE 



Scheme for Logging Observations — continued. 



Line. Symbol. 



.^10 

/n 

(J 
h 
i 

J 
k 
1 

m 
n 



P 
1 



dr 

dw 

w 

R 

I' 

T 

dr 
dw 

dt 

dw 



a -s 
C 



W„,o 



Explanation. 



Added weight w.'k, in grams : — Scale reading Rjo in millimetres. 

Mean increment of immersion produced by the addition of O'l gram to the external load. 

Mean added weight. 

Mean scale reading. 

Final temperature of liquid. 

Time when experiment was finished. 

Mean of initial and final temperatures. 

Difference of mean reading K from 50 mm. (50 -R). 

Weight which immerses dr millimetres of stem in the liquid. 

ib + chor : Added weight which produces immersion up to 50 mm. at T° C. 

= T - T : Departure of mean temperature from the standard temperature. 

Weight to be added to compensate the displacing value of dt. 

= w + dWi- + dwi : Total mean added weight required to immerse the closed hydrometer in the liquid up to 

50 mm. at the standard temperature, T. 
Weight of air contained in the open hydrometer, less that of the air displaced by the exposed part of the stem. 
= W+''i3;': — Total weight of liquid displaced by closed hydrometer when immersed in it up to 50 mm. at 

standard temperature, T. 
= W + Avt + a : Total weight of liquid displaced by open hydrometer when immersed in it up to 50 mm. at 

standard temperature, T. 
Total weight of distilled water displaced by hydrometer when immersed in it up to 50 mm. at standard 

temperature, T. 



"W + rwt 



for closed ' 
hydrometer. | 



W + ,M( + a for open 
W^;^ hydrometer. 



Specific gravity of the liquid at the standard temperature, T, referred to that of 
distilled water at the same temperature as unity. 



Numerical Examples in the Case of 

(a) Distilled Water at 19'5" C. 

(b) 7'0 gram-molecule Solution of Rubidium Chloride. 





(«) 


w 




{a) 


(6) 


a 


Hydrometer A. 


Hydrometer A. 


9 


10-69 mm. 


7 '33 mm. 


h 


136-11132 grams. 


184-67847 grams. 


h 


0'77 gram. 


14^75 grams. 


c 


February 10th, 1910. 


July 3rd, 1911. 


I 


51-39 mm. 


51^65 mm. 


d 


10.15 a.m. 


11.15 a.ra. 


7 


19-50°C. 


19-50° C. 


e 


19-50°C. 


19-50° C. 


k 


10.26 a.m. 


11.32 a.m. 


u 


'37 grams. 8-5 mm. 


14-15 grams. 7-5 mm. 


I 


19-50° C. 


19-50° C. 


U 


0-47 ,, 19-0 „ 


•25 ,, 14-2 ,, 


m 


1-39 mm. 


1 '65 mm. 


U 


0-57 ,, 30-0 ,, 


•35 ,, 212 ,, 


n 


0-0130 gram. 


02251 gram. 


U 


0-67 „ 40-5 ,. 


•45 „ 29^0 „ 





0^7570 ,, 


14-72749 grams. 


U 


0-77 „ 51-5 ,, 


•55 ., 36^7 „ 


V 


none 


none 


u 


0-87 ,, 62-5 „ 


•65 ,, 44-1 ,, 


9 


none 


none 


fn 


0-97 ,, 73-0 ,, 


•75 ,, 52^2 ,, 


r 


0^7570 gram. 


14-72749 grams. 


h 


1-07 ,, 83-5 ,, 


•85 ,, 59^3 ,, 


s 


0-13508 ,, 


0-13028 ,, 


ft 


1-17 „ 94-0 „ 


•95 ,, 66-8 „ 


t 


.1. 




f^o 




15-05 ,, 74-4 ,, 


u 


... 


199-53624 grams. 


fv 




-15 ,, 81-5 ,, 


V 


137-00340 grams. 


137-00340 ,, 


fu 


... 


■25 ,, 89-1 ,, 


w 


... 


1-456433 


As 


... 


-35 „ 95-5 ,, 









The weight 137 '0034 grams entered in line v includes that of the air contained in 
the hydrometer. Its value is obtained in the following manner : — 

As the result of many determinations, of which the example in § 84 is an 
instance, the mean added weight necessary to immerse the hydrometer to 50 mm. at 
19*50° C. was found to be 075471 gram. 

Hence weight of hydrometer and added weight = 136'86603 grams. 

Density of distilled water at 19-50° C. - 0-99834. 

1 r T -11 1 Til 136*86603 

Ihereiore volume oi distilled water displaced = o-QQft^A 



= 137*093 c.c. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 169 

By subtracting from this the volume of glass and shot ( ^ 24*6 c.c. ; see § 83), the 
resultant volume, 112'493 c.c, is that of the enclosed air. 

(The internal volume of the stem above the 50-mm. division is here disregarded.) 
The weight of this volume of air is obtained as follows : — 

Weight of 1 c.c. air under the atmospheric conditions during the 

experiments = 0"001208 gram. 
AVeight of 112-493 c.c. air = 0-13592 gram. 

These numbers give the amount of the air contained in the hydrometer when it 
carries an internal load of 95 '6 grams of lead shot. If this load is altered, tlie residual 
volume of air experiences a corresponding alteration. 

§ 87. Coy^rection for the rion-immersed Portion of Stem. — When the hydrometer is 
floating at 50 mm. in distilled water, there is a length of stem of 75 mm. in air — 
namely, 50 mm. to the end of the scale, and 25 mm. to the open end of the stem. 

By line g of the table in § 86, we see that 10 "6 9 mm. of scale are immersed by 0"1 
gram, and 75 mm. are immersed by 0*7 gram, whence the volume of the non-immersed 
portion of the stem may be taken as 0"7 c.c. 

By the Archimedean principle the non-immersed portion of the stem displacing this 
volume of air loses weight equal to that of the air so displaced ; so that, were the air 
removed from the surface of the liquid, the hydrometer would sink into the liquid and 
the scale reading would be higher. The value of this difference of scale reading is the 
weight of air displaced by the non-immersed portion of the stem. 

The weight of 0'7 c.c. of air, under the atmospheric conditions quoted above, is 
0-00084 gram. 

We have then, for the total weight which immerses the hydrometer to the 50-mm. 
division : — 

Weight of loaded hydrometer in vacuo =136'11132 grams. 
Weight of enclosed volume of air = 0-13592 

Added weight to immerse stem to 50 mm. = 0-75471 



137-00195 
Correction for exposed portion of stem = — 0*00084 



Sum= 137-00111 

This number represents the weight of distilled water displaced by the hydrometer 
up to the 50-mm. division when immersed in it at 19-5° C. 

The determination of the weight of any experimental solution displaced by the 

hydrometer up to the 50-mm. division is determined in a precisely similar manner. If 

any adjustment of the internal load of the hydrometer is made, its volume must be taken 

into account in estimating the volume of enclosed air. 

§ 88. The degree of accuracy attainable by the use of the hydrometer is best 
TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 22 



170 



MR J. Y. BUCHANAN ON THE 



illustrated by quoting the results of five series of observations, each series consisting 
of eleven independent observations made in a solution of calcium chloride containing 
6*3 gram-molecules of OaCl2 in 1000 grams of water. The table includes, by the 
method of least squares, the estimation of the probable error ^' of a single observation, 
and 7'o that of the arithmetical mean of each series. 

It will be seen from the example given in § 86 that we obtain a series of differences 
between the consecutive readings corresponding to added weights of 0"1 gram, and, 
to take the first case, 0"1 gram immerses 10"50 mm. of stem, an added weight of 0"0095 
would be required to immerse 1 "0 mm. of stem. 

When a series of readings has been made in the experimental solution with the 
hydrometer whose constants are known, the weight of solution displaced to a given scale 
division, which is one of the actual readings, is known, and the total weight of the 
hydrometer when floating at the same division in distilled water is obtained. With 
these data the calculation of the specific gravity of the solution is made. 

An example will illustrate this method : — 

Taking the first reading in series No. 1, using hydrometer A : 

1st Heading — 

4'0 grams added weight immersed 13*9 mm. 

Corrected weight of hydrometer =188"53748 grams. 

Volume of enclosed air == 108*56 c.c. 

Weight of enclosed volume of air = 0*12880 ,, 

Added weight = 4*00000 „ 



192-66578 
Correction for exposed stem (weight of 0*9 c.c. air) = — 0*00106 



Weight of solution displaced to 13*9 mm. 



= 192-66472 



Weight of distilled water displaced to 13*9 mm. 
Corrected weight of hydrometer 
Weight of enclosed volume of air 

Added weight to immerse 13*0 mm. 

•9 mm. 

Stem correction (weight of 0*9 c.c. air) 



= 136 
= 


•11132 
*13592 


136 
= 
= 


•24724 
41042 
•00837 


136 
= -0 


66603 
00110 



Weight of distilled water displaced to 1 3 9 mm. = 136*66493 

c, .n . 192*66472 

Specific gravity = ^gg.^^^gg = 1 "409760. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 171 



The table in § 90 includes three series obtained with hydrometer A, and two obtained 
with hydrometer B. The values of the mean specific gravity (S) furnished by each 
series, and its probable error (±^"o), expressed in units of the sixth decimal place, are 
collected in the following table : — 



\ Hydrometer. 

j 


s. 


±ro. 


A 


1-409752 


3-1 




1-409746 


3-5 




1-409746 


3-6 


B 


1-409727 


4-5 




1-409753 


6-4 



It will be seen that the uncertainty of the means of each series lies entirely in the 
sixth decimal place. The mean of the five means tabulated is 1*409744, and its 
probable error is ±3 '16 in the sixth decimal place. 

§ 89. In order to reap the full benefit of the precision of which the hydrometric 
method is capable, the operations must be carried out with attention to every pre- 
caution, and the experimental data must be recorded according to strict method. 

Scrupulous cleanliness is of the first importance, and the operations must be carried 
out with attention to all the precautions usually observed in laboratories from which 
exact work is expected to proceed. 

It is important that the room in which the observations are made should have a 
north light and be entirely under the control of the experimenter, who is its only 
occupant. This is essential, because the management of the temperature of the room, 
which must be that which the experimenter has found by his own experience to be the 
one which maintains the experimental liquid constantly at the selected standard 
temperature while the observations are being made, is the most important element of 
success and the most diflicult of achievement. The conditions are similar to those 
which have to be observed in the room in which gas analysis is made by Bunsen's 
original method, only they are rather more stringent. For myself, when I begin 
hydrometric observations I always lock the door, a practice which I adopted on board 
the Challenge?' and have adhered to ever since. 

The conclusion arrived at from the discussion on temperature conditions which 
is given in Section IV. on the closed hydrometer applies with equal force in the 
use of the open hydrometer. An interesting diff"erence occurs in experiments on 
strong solutions, since they have a lower specific heat, which may fall as low as 0*5, 
as in the case of most concentrated solutions of CaCl2, so that the thermal mobility 
of these solutions is greater, and this condition may be met by allowing a somewhat 
increased margin of difierence between air and solution temperature when the 
compensating luminous flame is used. 



172 



MR J. Y. BUCHANAN ON THE 



§ 90. Table of Specijic Gravities calculated from Single Observations made with 
Hydrometers A and B when floating hi a Solution of Calcium Chloride con- 
taining 6*3 gram-molec\des in 1000 grams of Water. 







Hydrometer A. 


Added 
Weight in 
Grams for 

Hydro- 
meter B. 


Hydrometer B. 


Added 
Weight in 
Grams for 

Hydro- 
meter A. 


Series 1. 


Series 2. 


Series 3. 


Series 1. 


Series 2. 


Scale 

Reading 

in 

Milli- 
metres. 


Specific 
Gravity 
calculated 
from Single 
Observa- 
tions. 


Scale 

Reading 

in 

Milli- 
metres. 


Specific 
Gravity 
calculated 
from Single 
Observa- 
tions. 


Scale 

Reading 

in 

Milli- 
metres. 


Specific 

Gravity 
calculated 
from Single 

Observa- 
tions. 


Scale 

Reading 

in 

Milli- 
metres. 


Specific 

Gravity 
calculated 
from Single 
Observa- 
tions. 


Scale 

Reading 

in 

Milli- 
metres. 


Specific 
Gravity 
calculated 
from Single 
Observa- 
tions 


4-0 
4-1 
4-2 
4-3 
4-4 
4-5 
4-6 
4-7 
4-8 
4-9 
5-0 


13-9 
21-8 
29-6 
37-2 
44-6 
52-2 
59-8 
67-1 
74-7 
82-2 
89-7 


1-409760 
735 
731 
730 
744 
749 
762 
776 
764 
762 
761 


141 

21-7 
29-7 
37-2 
44-9 
52-1 
59-8 
67-2 
74-9 
82-4 
89-9 


1-409741 
745 

722 
730 
722 
765 
768 
772 
750 
748 
747 


14-0 

21-5 
29-5 
37-4 
44-8 
52-1 
60-1 
67-5 
74-8 
82-6 
89-7 


1-409750 
764 
741 
711 
731 
765 
741 
744 
760 
730 
767 


2-05 
2-15 
2-25 
2-35 
2-45 
2-55 
2-65 
2-75 
2-85 
2-95 
3-05 


7-1 
15-9 
24-5 
33-5 
42-8 
50-8 
60-0 
67-9 
77-8 
86-0 
94-8 


1-409740 
732 
744 
725 
683 
751 
723 
725 
694 
740 
739 


7-0 
15-8 
24-7 
33-8 
42-0 
50-9 
59-0 
68-1 
76-9 
86-0 
94-8 


1-409750 
742 
724 
696 
760 
742 
818 
790 
782 
740 
739 


Mean Specific 
Gravity. 


§ 


1-409752 




1-409746 




1-409746 






1-409727 




1-409753 


Probable 
error, ex- 
pressed 
in units 
of the 6th 
decimal 
place 


of a single 
observation 
±r. 


10-3 




11-6 




11-9 




14-8 




20-1 


of the arith- 
metical 
mean +?'q. 


31 




3-5 




3-6 






4-5 




6-4 



The hydrometer A, constructed on the above specification, has proved itself, in use, 
to be an excellent model. Its volume, about 137 c.c, is very suitable, being sufficiently 
great to secure precision without rendering it necessary to use extravagant quantities 
of very soluble salts, which, in the case of costly preparations, might be prohibitive. 
It is very steady, and this is principally due to the fact that the ballast is all contained 
in the bulb at the lower extremity. The position of the centre of gravity of the 
instrument is thus kept very low, and in the spherical bulb the ballast cannot shift. 



Section XIII. — On the Specific Gravity and Displacement of Solutions of 
Salts of the Ennead MR which have nearly the same Molecular 
Weight and may be looked on as "Isomeric." 

§ 91. There are three such groups of salts in the ennead, namely, KBr and RbCl ; 
KI, RbBr, and CsCl ; and Rbl and CsBr. Experiments have been made on strong 
solutions of the first group, KBr and RbCl. To these " natural isomers " we have added 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 173 

an (artificial) isomer consisting of a mixture which contains KCl and KI in equal mole- 

Cl + I 
cular proportions, so that it may be represented by the formula K — - — '- This mixture 

contains 3 TOO per cent, of KCl and 69 "00 per cent. KI. 

It was found that at 19'5°C., the temperature used in these experiments, 
702'25 grams of this mixture saturated 1000 grams of water. No more could be 
dissolved without leaving a residue. This amount was made up of 217*65 grams of 
KCl and 484*60 grams KI, representing 2*917 gram-molecules of each salt. From the 
tables of solubility of these well-known salts we find that 217*65 grams of KCl 
saturate 632*26 grams of water at 19*5° C. ; and if we imagine that this quantity of 
water is wholly taken possession of by 217*65 grams of KCl, there remains 367*74 
grams of water to accommodate the 484*60 grams KI. But, at 19*5° C, 367*74 grams 
of water require 530*28 grams KI to produce saturation. Therefore, though 
saturated with the mixture, the 1000 grams of water is not saturated with both the 
individual salts. 



92. Table giving Results of Specific Gravity Determinatio7is made ujwn Solutions 
of Rubidium Chloride, Potassium Salt of mixed Halides, and Potassium 
Bromide, of different Concentrations. 



m. 
1. 



W. 

2. 



S. 
3. 



rfS 
d')n 



log A. 
.5. 



d log A 
dm 





rfA 


V 


A 








dm 


m, 


7. 


8. 


9. 



log A„ 



log Aj - 3 



10. 



EbCl= 121-0. T= 19-50° C. 



7" 


1847-000 


6 


1726000 


5 


1605-000 


4 


1484-000 


3 


1363-000 


2 


1242-000 


1 


1121-000 



•456464 
-406075 
•351760 
•292983 
•229284 
■159851 
-083782 



0050389 
0054315 
0-058777 
0-063699 
0-069433 
0-076069 



3-1031672 
3-0890324 
3-0745755 
3-0598410 
3-0448789 
30297194 
3-0146637 



0-0141348 
0-0144569 
0-0147345 
0-0149621 
0-0151595 
0-0150557 



1268-140 
1227-531 
1187-341 
1147-733 
1108 865 
1070-827 
1034-341 



40-609 
40-190 
39-608 
38 868 
38-038 
36-486 



38-306 
37-922 
37-468 
36-933 
36-288 
35-413 
34-341 



7-035 
6-072 
5-086 
4081 
3060 
2-027 
1-000 



K^l±^ = 120-35. 



T = 19-50° C. 



1601-75 
1481-40 
1361-05 
1240-70 
1120-35 



-336904 
-280510 
•219453 
-152989 
-080182 



0-056394 
0-061057 
0-066464 
0-072807 



3-0784945 

3-0632893 0-0152052 

3-0477089 00155804 

3-0318418 0-0158671 

3-0158567 0-0159851 



1198-104 
1156-880 
1116-115 
1076-073 
1037-186 



41-224 

40-765 
40-042 
38-887 



39-620 
39-220 
38^705 
38036 
37-186 



4-950 
3-991 
3-009 
2-008 
1-000 



KBr= 119-1. T= 19-50° C. 



1595-500 


1-343255 




1476-400 


1-285584 


0-057671 


1357-300 


1-223113 


0-062471 


1238-200 


M55257 


0-067856 


1119-100 


1^081211 


0-074046 



3-0747383 
3-0601036 
3-0452092 
3-0301122 
3-0149585 





1187-786 




37-557 


0-0146347 


1148-428 


39-358 


37-107 


0-0148944 


1109-709 


38-719 


36-569 


0-0150970 


1071-796 


37-913 


35-898 


0-0151537 


1035-043 


36-753 


35-043 



4-996 
4-018 
3 022 
2-013 
1-000 



174 



MR J. Y. BUCHANAN ON THE 



§ 93. Solubility. — The molecular solubility of each of these salts, that is, its solu- 
bility expressed in gram-molecules salt per thousand grams of water at 19'5° C, is : — 



Salt. 


RbCl. 


^01 + 1 
^ 2 • 


KBr. 


Molecular weight . . 
Gram-molecules in 1000 grams water . 


121 

7-77 


120-35 

5-83 


119-1 
5-7 



The discussion will, however, be confined to the relation of solutions of these salts 
which contain, per thousand grams of water, 5 or a smaller number of gram-molecules 
of salt. 

Considering the change of specific gravity over a range of concentration varying 
from 5 to 1 gram-molecules per thousand grams of water, that of rubidium chloride 
varies from i-351760 to r083782, that of the potassium salt of the mixed halides 
from 1-336904 to 1-080182, and that of potassium bromide from 1-343255 to 
1*081211. Although potassium bromide has the lowest molecular weight, there is a 
closer agreement in the specific gravity of the solutions of this salt with those of 
rubidium chloride than with those of the potassium salt of the mixed halides. At the 
same time the nature of the change of values with change in concentration of solution, 
as indicated by the numbers representing the difi'erences of consecutive specific gravities 
given in the column d^/dm of the tables, shows that in this respect the potassium salts 
exhibit a closer relationship among themselves than either of them does with rubidium 
chloride. Also the nature of the decrease in this value with increasing concentration 
seems to indicate the fact that the increase in specific gravity becomes more nearly 
proportional to the increase in concentration in the strongest solutions. 

It will be observed that the actual weight of each salt per thousand grams of 
water in each solution is slightly diff'erent, and if the specific nature of the salts were 
the same in each of these solutions, and the masses of them present in the solution 
were equal, as their molecular weights would in that case be, the specific gravities of 
these solutions and the constants derived from them would be dift'erent from those in 
table § 92. 

§ 94. For comparison in this sense the specific gravities have been adjusted to the 
value which they would have if their gram-molecules had the uniform weight 121, 
which is the actual molecular weight of the heaviest of the three, namely, rubidium 
chloride. 

The following table gives the specific gravities adjusted in this sense :— 



1 

Multiples of 121 grams of Salt per 1000 grams Water . 


5. 4. 

1-351760 1-292983 
1-338724 1-282025 
1-3487311 1-290140 

1 


3. 


2, 1. 


Observed specilic gravities for solutions of RbCl . 
Oalculated „ „ „ K „ 
» » .. >, K:Br . 


1-229284 
1-220131 
1-226672 


1-159851 
1-153815 
1-157734 


1-083782 
1-080575 
1-082507 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 175 

It will be observed that the adjustment of the molecular weight does not materially 
aflfect the relations of the solutions as regards their specific gravities. 

Returning to the consideration of the data obtained from the original experiments, 
the comparison of the displacements for the same concentrations in the case of solutions 
of each of the three salts shows a close agreement to exist between the values for 
rubidium chloride (which are lower in each case) and potassium bromide, while the 
values for the potassium salts of the mixed halides stand quite apart and are much 
higher than the corresponding values for the other two salts. 

If we compare the differences of displacements of equivalent solutions of RbCl and 
KBr for m = 5, thenARbci- Akbi = 0'445, and for m=l it is 0'702. The differences 



of displacements between corresponding solutions of K 



Cl + I 



and KBr are, for m = 5, 



Ag.ci+1- AKBr= 10-318, and form= 1, A^ci+i- AKBr = 2-143. 



The molecular displacement of each of these salts in crystal, as given in 

.01 + 1 



127, is 



RbCl = 44-710, K^ 



46-406, KBr = 44-460, 



and if the sum of the displacements of the constituent materials forming the solution, 
i.e. 1 gram-molecule and 1000 grams of water, are compared with the displacement of 
the solution obtained from the constituents, the following results are arrived at : — 





RbCl. 


2 


KBr. 


Sum of displacement of constituents 
Displacement of solution 

Difference 


1044-710 
1034-341 


1046-406 
1037-186 


1044-460 
1035-043 


10-369 


9-220 


9-417 



Here, with regard to the change in displacement when solution is effected, the 
potassium salts are quite comparable, while the rubidium chloride shows a much greater 
change, although in the case of values for the sum of the displacements of the con- 
stituents, rubidium chloride and potassium bromide are the more comparable. 

§ 95. Difference of Displacement, dA. — c?A gives the increment of displacement 
of a mass of 1000 grams of water produced by successive additions of 1 gram-molecule 
of salt to that already in solution. 

The values of v/m represent the mean increment of displacement of 1000 grams of 
water per gram-molecule of salt when w gram-molecules have been dissolved in it. 

The values of dA from 1 to 4 gram-molecules for each salt show that, with the 
exception of that for the 1 gram- molecule, they are lowest in the case of potassium 
bromide, and the values for corresponding concentrations of rubidium chloride very 
closely approximate to them, while those for the potassium salt of the mixed halides 
show a considerable divergence. 



176 MR J. Y. BUCHANAN ON THE 

Thus the value of d^ for a 4 gram- molecule solution of potassium bromide, 
which is 39'358, diminishes to 36"753 for the 1 gram-molecule solution, while that 
for rubidium chloride diminishes from 39'608 to 36*486 for the same range of con- 
centration, and for the potassium salt of the mixed halides the two values are 41 '224 
and 38-887. 

If we express these pairs of values as ratios, the following values are obtained : — 

RbCl. K^l+i. KBr. 

2 

Actual values = 39-358: 36-753 41-224 : 38-887 39-608:36-486 

Ratio = 1:0-9212 1:0-9433 1:0-9338 

While, therefore, the values of (iA closely approximate in the case of rubidium 
chloride and potassium bromide, yet the ratios given above show that the rate of 
decrease in the value of d^ for potassium bromide lies between those for the other 
salts, but closer to that for the potassium salt of the mixed halides. 

An inspection of the values of vjm shows that all the values for RbCl are lower 
than the corresponding values for the other two salts, although the values for potassium 
bromide approach very close to them, while those for the potassium salt of the mixed 
halides are much higher. 

The values of vjm for the 5 and 1 gram-molecule concentrations for each of the 
salts are 37-468 and 34-341 for rubidium chloride, 37-557 and 35-043 for potassium 
bromide, and 39-620 and 37-186 for the potassium salt of the mixed halides; and 
expressed as ratios, as in the case of the values for c?A, we have for 



RbCl. 


2 


KBr. 


Actual values = 37-468 : 34-341 


37-557:35-043 


39-620:37-186 


Ratio = 1 : 0-9165 


1 : 0-9386 


1 : 0-9331 



Here, the similarity of the ratios for the potassium salt of the mixed halides and 
potassium bromide shows a similar rate of decrease of this value vjm for these two salts, 
while that for rubidium chloride shows a considerable departure from either of them. 

The agreements which exist, when the values ^ — and — for the three salts are 

dm m 

compared, seem to indicate that the molecules of rubidium chloride and potassium 

bromide exert almost equal effects in the displacement of solution, but the nature of 

change of displacement with change of concentration shows that the potassium salts 

are more allied in this respect. 

§ 96. The Displacement of Solutions of the Potassium Salt of the Mixed Halides 
when considered in reference to the Displacement of Solutions of the Constituent 
Salts. — The following table gives full data relating to the displacement, difference of 
displacement, and mean increment of displacement of solutions of different concentra- 
tions of potassium chloride, potassium salts of the mixed halides, and potassium iodide. 

The experiments were made with the open hydrometers A and B (see § 82) at 
the constant temperature 19-50° C. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 177 



Concen- 


KCl. 




gCl-fl 
2 






KI. 




tration 
in grain- 
raolecules 
















Displace- 


Difference 
of Dis- 


Mean Incre- 
ment of 


Displace- 


Difference 
of Dis- 


Mean Incre- 
ment of 


Displace- 


Difference 
of Dis- 


Mean Incre- 
ment of 


per 1000 
grams of 


ment. 


placement. 


Displacement. 


ment. 


placement. 
dA 


Displacement. 

V 




placement. 
d£. 


Displacement. 

V 


Water. 


A. 


dm 


m 


A. 


dm. 


m 


A. 


dm 


m 


5 








1198-104 




39-620 


1240-334 




48-067 


4 


1123-005 




30-751 


1156-880 


41-224 


39-220 


1190-788 


49-546 


47-697 


3 


1090-533 


32-472 


30-178 


1116-115 


40-765 


38-705 


1141-787 


49-001 


47-262 


2 


1059-617 


30-916 


29-808 


1076-073 


40-042 


38-036 


1093-384 


48-403 


46-692 


1 


1028-904 


30-713 


28-904 


1037186 


38-887 


37-186 


1046-189 


47195 


46-189 


1/2 


1014-001 


29-806 


28-002 


1018-343 


36-308 


36-686 


1022-778 


46-822 


45-556 



The composite salt was prepared by mixing the component salts KCl and KI in 
the proportion of their molecular weights, 74*6 : 166"1, so that 1 gram-molecule of 
the salt, which weighs 120 '35 grams, would contain ^ gram-molecule of each of the 
salts, namely 37'3 grams of KCl and 83'05 grams of KI. When 1 gram-molecule 
of this salt is dissolved in 1000 grams of water, the displacement of the resultant 
solution may be compared with the mean of the displacements of the solutions 
KCl + 1000 grams of water and KI -|- 1000 grams of water. 

The following table gives the results of such a comparison : — 

§ 97. Table of Values of Displacements of Solutions of Potassium Salts of the 
Mixed Halides which have been obtained — 

A, by experiment. 

B, calculated by the method detailed above. 

B — A, the difference of calculated and observed results. 



Concentration in 

gram-molecules per 

1000 grams Water. 

m. 


A. 


B, 


B-A. 


5 

4 
3 

2 
1 

1/2 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 


1198-104 
1156-880 
1116-115 
1076073 
1037-186 
1018-343 
1009-071 
1004-474 
1002-213 
1001-122 
1000-535 
1000-262 
1000-117 
1000-044 


1156-896 
1116-160 
1076-500 
1037-546 
1018-389 
1009-090 
1004-495 
1002-228 
1001-118 
1000-559 
1000-282 
1000-133 
1000-076 


0-016 
0-045 
0-427 
0-360 
0-046 
0-019 
0-021 
0-015 
-0-004 
0-024 
0-020 
0-016 
0-032 



This table shows that, with the possible exception of the 1/32 gram-moleeule 

solution, the mixture of wKCl-f 1000 grams of water and mKI-|- 1000 grams of water 

is accompanied by contraction. 

TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 23 



178 MR J. Y. BUCHANAN ON THE 

Section XIV. — The Specific Gravity and the Displacement of Solutions 
OF THE Chlorides of Beryllium, Magnesium, and Calcium. 

§ 98. For the purpose of determiniDg these constants, hydrometric observations 
were made on solutions of the three salts the concentrations of which ranged from 1/2 to 
1/1024 gram -molecule per thousand grams of water, the experiments being made with 
the closed hydrometers Nos. 3 and 17. Experiments were also made on strong solutions 
of calcium chloride and magnesium chloride, using the open hydrometers A and B ; 
the concentrations of these solutions varied from 1 gram-molecule per thousand grams 
of water to the highest attainable degree of supersaturation. It was when the experi- 
ments on a supersaturated solution of calcium chloride were in progress that the 
observations were made which revealed the remarkable state of unrest in that solution 
which preceded its partition into crystals and mother-liquor with liberation of heat. 
The details of this experiment are given in Section XV., and from them it will be 
seen that the range of supersaturation which can be explored hydrometrically when the 
salt in solution is chloride of calcium is considerable. The solution of magnesium 
chloride which is saturated at 19"5°C. contains 5*918 gram-molecules (564*123 grams) 
of MgClg in 1000 grams of water. A supersaturated solution, containing 5*982 gram- 
molecules of salt per thousand grams of water was cooled to 16'5° C, at which tempera- 
ture the saturated solution contains 5*853 MgClg per thousand grams of water; yet, 
with this small degree of supersaturation, the slightest disturbance, such as lifting the 
beaker, induced crystallisation in the solution. This shows that the limits of super- 
saturation are restricted, and that it would certainly be discharged by an attempt to 
make hydrometric observations in the solution. 

This difference in the behaviour of the supersaturated solutions of these two salts is 
interesting. On the one hand we have the calcium chloride, which produces a high 
degree of supersaturation with great absorption of heat, and offers great resistance to 
crystallisation ; while magnesium chloride can produce solutions attaining only to a 
moderate degree of supersaturation with very moderate absorption of heat, and the salt 
crystallises from such solutions on the slightest provocation. With a view to a com- 
parison with the thermal behaviour of chloride of calcium, some observations were made 
on the heat of solution of magnesium chloride in water, while experiments were being 
made to determine the concentration of solutions saturated with the salt at different 
temperatures. It was found that when the quantities of the crystallised salt 
MgClgjBHgO and water used were such as to produce a solution the concentration of 
which was about 2*0 gram-molecules of MgClg per thousand grams of water, the 
dissolution of the salt was accompanied by an appreciable liberation of heat. When 
the conditions of the experiment were such that a saturated solution was formed, some 
of the crystals remaining undissolved, the dissolution of the salt was accompanied by 
absorption of heat. When the saturated solution was produced by fractions, it was 
observed that during the dissolution of the first fraction the thermometer indicated a 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 179 



99. 



Table giving Specific Gravity Values obtained from Experiments made upon 
Solutions of the Chlorides of Beryllium, Magnesium, and Calcium. 











cZlog A 
dm 






V 


logA™-3_^ 




TO. 


W. 


S. 


log A. 


A. 


rfA. 


in 


log Aj - 3 


x-m. 


1. 


2. 


3. 


4. 


5. 


6. 


7. 


8. 


9. 


10. 






BERYLLIUM CHLORIDE = 


BeCl2 = 80-00. T = 19-50°C. 






1/2 


1040-0000 


1-025620 


3-0060468 




1014-0208 




28-04 


1 

X 

2-000 


m X 
0-000 


]/4 


1020-0000 


1 1-013055 


3-00-29671 


0-0123188 


1006-8555 


7-1653 


27-42 


4-076 


-0 076 


1/8 


1010-0000 


1-006599 


3-0014648 


0-0120181 


1003-3785 


3-4770 


27-03 


8-256 


-0-256 


1/16 


1005-0000 


1-003347 


3-0007149 


0-0119995 


1001-6745 


1-7310 


26-37 


16-916 


-0-916 


1/32 


1002-5000 


1-001587 


3-0003957 


0-0102144 


1000-9115 


0-7360 


29-17 


30-562 


+ 1-438 


1/64 


1001-2500 


1-000742 


3-0002204 


0-0112192 


1000-5076 


0-4039 


32-49 


54-870 


+ 9-130 


1/128 


1000-6250 


1-000325 


3-0001302 


0-0115430 


1000-2999 


0-2077 


38-39 


92-872 


+ 35-128 


1/256 


1000-3125 


1-000083 


3-000i;996 


0-0078259 


1000-2295 


0-0704 


58-75 


121-362 


+ 134-638 


1/512 


10001562 


0-999957 


3-0000865 


0-0067225 


1000-1992 


0-0303 


101-99 


139-778 


+ 372-222 


1/1024 


10000781 


0-999906 


3-0000747 


0-0120832 


1 1000-1720 


0-0272 


176-13 


166-854 


+ 862-146 






M 


A.GNESIUM ( 


CHLORIDE = 


MgCl2=95- 


32. T = 19-50° C. 






5-9820 


1570-2050 


1-338895 


3-0692097 




1172-7345 




28-87 


X 

7-538 


X-m 
1-556 


5-918-2 


1564 1230 


1 336101 


30684316 


0-0121959 


1171-6623 


2-0722 


28-84 


7-453 


1-535 


5-5 


1524-2600 


1-317763 


3-0632219 


0-0124574 


1156-7032 


13-9591 


28-49 


6-886 


1 -386 


50 


1476-6000 


l-29o011 


3 0569894 


0-0124650 


1140-2220 


16-4812 


28 04 


6-207 


1-207 


4-0 


1381-2800 


l-246tJ66 


3-0445317 


0-0124577 


1107-9793 


32 2427 


26-99 


4-850 


0-850 


30 


1285-t)600 


1-193986 


3-0322282 


0-0123035 


1077-0310 


30-9483 


25-68 


3-510 


0-510 


2-0 


1190-6400 


1-136425 


3-0202398 


0-0119884 


1047-7070 


29-3240 


23-85 


2-204 


0-204 


1-0 


1095-3200 


1-072407 


3-0091814 


0-0110584 


1021-3660 


26 3410 


21-37 


1000 
1 


0-000 
1 1 


1/2 


1047-6600 


1-037385 


3-0042803 


0-0098020 


1009-9047 


11-4613 


19-81 


X 

2-145 


m X 
- 0-145 


1/4 


1023-8300 


1-019096 


3-0i)201:>7 


0-0090705 


1004 6453 


5-2594 


18-58 


4-562 


-0 562 


1/8 


1011-9150 


1-009675 


3-0009624 


0-0084024 


1002-2185 


2-4268 


17-75 


11-982 


-3-982 


1/16 


1005-9575 


1004f593 


3-0004598 


-0080422 


1001-0593 


1-1592 


16-95 


19-968 


-3-968 


1/32 


1002-9787 


1-002461 


3 0002242 


0-0075385 


1000-5164 


0-5429 


16-52 


40-948 


-8-948 


1/64 


1001-4893 


1-001244 


3-0001063 


0-0075411 


1000-2450 


0-2714 


15-68 


86-299 


-22-299 


1/128 


1"00 7446 


1-000639 


3-0000458 


0-0077516 


1000-1055 


0-1395 


13-50 


200-336 


-72-o36 


1/256 


1000-37-23 


1-000299 


3 0000318 


0-0035865 


1000-0732 


0-0323 


18-74 


288-542 


-36-542 


1/512 


1000-1861 


1-000172 


3-0000065 


0-0129587 


1000-0150 


0-0582 


7-68 


1410-353 


-898-353 


1/1024 


1000-0930 


1-000082 


3-0000047 


0-0017715 


1000-0110 


0-0040 


11-26 


1920-795 


-896-795 






C 


ALCIUM CH 


LORIDE = Ca 


"12 = 111-00. 


T = 19-50 


'C. 






6-627 


1735-620 


1-424183 


3-0858888 




1218-6775 




33-00 


X 

8-279 


x-m 
1-652 


6-613 


1734 043 


1-423500 


3-0857023 


0-0133214 


1218 1540 


0-5233 


32-99 


8-261 


1-648 


6-6 


1732-600 


1-422871 


3-0855327 


0-0130461 


1217-6787 


0-4753 


32-98 


8-245 


1-645 


6-5 


1721-500 


1-418572 


3-0840556 


0-0147710 


1213-5441 


4-1346 


32-85 


8-102 


1-602 


6-4 


1710-400 


1-414247 


3-0825725 


0-0148310 


1209-4069 


4-1372 


32-72 


7-959 


1 559 


6-3 


1699-300 


1-409741 


3-0811308 


0-0144170 


1205-39S8 


4-0081 


32-60 


7-820 


1-520 


6-2 


1688 200 


1-405270 


3-0796641 


0-0146670 


1201-3349 


4-0639 


32 47 


7-697 


1-497 


6-1 


1677-100 


1-400460 


3-0782883 


0-0137580 


1197-5352 


3-7997 


32-38 


7-546 


1-446 


6-0 


1666-000 


1-395919 


3-0768148 


00147350 


1193-4791 


4-0561 


32-24 


7-421 


1-421 


5-0 


1555-000 


1-342951 


3-0636673 


0-0131475 


1157-8899 


35-5892 


31-58 


6-165 


1-165 


4-0 


1444-000 


1-284536 


3-0508210 


0-0128463 


1124-1413 


33-7486 


31-03 


4-899 


0-899 


3-0 


1333-000 


1-227685 


3-0357431 


0-0150779 


1085-7832 


38-3581 


28-59 


3-445 


0-445 


2-0 


1222-000 


1-160139 


3-0225612 


0-0131819 


1053-3221 


32-4611 


26-66 


2-175 


0-175 


1-0 


1111-000 


1-084776 


3-0103741 


0-0121871 


1024-1745 


29-1476 


24-17 


1000 
1 


0-000 
1^ 1 


1/2 


1055-5000 


1-043739 


3-0048663 


0-0110155 


1011-2681 


12-9064 


22-54 


X 

2-132 


m X 
-0-132 


1/4 


1027-7500 


1-022253 


3-0023290 


0-0101489 


1005-3773 


5-8908 


21-50 


4-454 


-0-454 


1/8 


1013-8750 


1-011231 


3-0011340 


0-0095604 


1002-6146 


2-7627 


20-92 


9-148 


-1-148 


1/16 


1006-9375 


1-005660 


3-0005513 ' 


0-0093233 


1001-2703 


1-3443 


20-32 


18-817 


-2-817 


1/32 


1003-4687 


1-002763 


3-0003055 1 


0-0078656 


1000-7037 


0-5666 


22-52 


33-954 


-1-954 


1/64 


1001-7343 


1-001423 


3-0001349 ! 


0-0109145 


1000-3109 


0-3928 


19-90 


76-851 


-12-851 


1/128 


1000-8672 


1-000729 


3-0000599 


0-0096038 


1000-1380 


0-1729 


17-66 


173-017 


-45-017 


1/256 


1000-4336 


1-000378 


3-0000241 


0-0091699 


1000-0556 


0-0824 


14-23 


429-747 


-163-747 


1/512 


1000-2168 


1-000179 


3-0000163 


0039680 


1000-0378 


0-0178 


19-35 


632-953 


-120-953 


1/1024 


1000-1084 


1-000093 


3-0000066 


0-0099430 


1000-0154 


0-0224 


15-77 


1553-009 


-629-009 



180 MR J. Y. BUCHANAN ON THE 

rise of temperature. On adding further fractions of salt this ceased, the thermal effect 
was reversed, and when saturation had been effected the temperature of the solution 
had fallen below the initial temperature of the water used. 

No experiments have been made on solutions of beryllium chloride of greater con- 
centration than 1/2 gram-molecule per thousand grams of water. 

The preceding table contains all the experimental results and the deductions there- 
from. The form and the symbols used have been already explained. 

§ 100. Before discussing the data of the tables, attention must be directed to the 
distinctive characters of the three salts. While the bases BeO, MgO, and CaO give an 
alkaline reaction with litmus paper, the chlorides of magnesium and of calcium are 
neutral, while that of beryllium is acid. In order to obtain, if possible, a neutral 
solution of BeClg, the method adopted was to proceed by way of the sulphate and 
double decomposition with chloride of barium. A solution of pure crystallised sulphate 
of beryllium, containing exactly 1 gram-molecule of BeSO^ in 1000 grams of water, 
was made, and with it was mixed a quantity of a solution of barium chloride containing 
exactly 1 gram-molecule of BaClg in 1000 grams of water. The barium sulphate was 
precipitated completely, and the supernatant liquid contained exactly 1 gram-molecule 
of BeClg in 2000 grams of water, or, at the rate of 1/2 BeCla in 1000 grams of water. 
This solution, which still had an acid reaction, was used for the preparation of the less 
concentrated ones by exact dilution. It is impossible to produce solutions in this 
way for which m>l/2, on account of the bulk of the barium sulphate produced. 
Solutions of the highest attainable degree of concentration were prepared in the case of 
magnesium chloride and calcium chloride, the concentration being determined by the 
usual chemical methods, and the solutions of a lesser degree of concentration were 
prepared from these. 

In all cases the solutions were prepared by diluting the more concentrated solution 
immediately preceding it, a method capable of a high degree of precision, which is 
shown by the fact that, after the experiments on strong solutions of MgCl.2 were com- 
pleted, a single determination of the concentration of the 1 gram-molecule solution of 
MgClz gave a result of 0'9991 gram-molecule of salt in 1000 grams of water. The 
result was obtained by a determination of the chlorine content. 

Chlorine found ....... 662 per cent. 

,, calculated ...... 6"63 „ 



Diff. 0-01 

The specific gravity experiments were carried out by the use of the open hydro- 
meters A and B in the case of strong solutions of the salts, magnesium chloride and 
calcium chloride ; and by the use of the closed hydrometers Nos. 3 and 17 in the case 
of the solutions of each of the three salts where the concentrations were less than I'O 
gram-molecule of salt in 1000 grams of water. 

The constant experimental temperature was 19 "50° C. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 181 

Concentration of the Solutions (m). — The highest concentrations were those of 
slightly supersaturated solutions in the case of magnesium and calcium chlorides. The 
solution of magnesium chloride which is saturated at 19 '50° C. contains 5 "91 82 gram- 
molecules of salt in 1000 grams of water, and is the second of the series of strong 
solutions of this salt. The solution containing 5*5 gram-molecules of salt was prepared 
from this solution, and then solutions were experimented on having a common difference 
of 1 gram-molecule, and ranging from 5'0 to TO gram-molecule. 

The 6 '6 27 gram-molecule solution of calcium chloride is supersaturated, and the 
6 "6 13 gram-molecule forms a solution which is saturated with CaClg at 19 "50° C. This 
solution is of interest as being the mother-liquor obtained after crystallisation from the 
solution which showed the condition of unrest described in Section XV. The experi- 
ments on this solution were made at a much earlier date than the others included 
in this table, but the results are included here in order to give a complete list of the 
experiments made. 

As will be seen, the solutions for which w = 6'6 to 6-0 decrease regularly in 
concentration by 0*1 gram-molecule. They were made in order to trace the changes of 
displacement due to small changes of concentration in nearly saturated solutions. For 
m = 6'0 to 1"0 gram-molecules the common difference of concentration of consecutive 
solutions is 1 "0 gram-molecule. 

Discussion of Results. Specific G7'cwity. — With regard to the agreement of 
the individual results among themselves for a particular series, the analysis of the 
specific gravity results in the case of the calcium chloride solution containing 6*3 gram- 
molecules of salt in 1000 grams of water (see § 90) affords a fair criterion, and the 
usual number of series of observations made for each concentration was six, three with 
each hydrometer. In all the experiments the results of which are included in these 
tables, the temperature of the solution remained constant at 19 "50° C. Comparing the 
specific gravities of solutions of the three salts, having the same molecular concentration, 
and m being less than 1 "0, the values in all cases increase with increasing molecular 
weight. Thus, for m=l/2 the specific gravities of the solutions rise from 1*025620 for 
BeClg to 1-037385 for MgClg and to 1-043739 for CaCl2, and the same feature is 
observed in comparing the values for all concentrations down to w = 1/1024. 

It is interesting to compare the increments of specific gravity (S— 1), (which for 
this purpose are conveniently multiplied by 1000), of the solutions with the molecular 
weights of the salts dissolved in them. We have them in the following table, for 
solutions for which m = l/2 : — 

MR = BeCl.^. MgCIa. CaCls. 

1/2 MR = 40 47-66 55-5 

1000 (S-1) = 25-620 37-385 43739 

^T^f.7.^^ = 0-640 0-784 0-784 

1/2 MR 

When w = l/16 and 1/128 we have the following values : — 



182 MR J. Y. BUCHANAN ON THE 

MR = BeClj. MgClo. CaClg. 

- ^^,^,?i^.'^ - = 0-669 0-821 0-815 

1/16 MR 

^??^ ^^,~t!^ = 0-520 0-858 0-841 

1/128 MR 

From this table we see that the increment of specific gravity produced by dissolving 
1/2 ME. in 1000 grams of water is exactly proportional to the molecular weight of the 
salts in the case of MgClg and CaCla, and that this proportionality is maintained for 
values of ??i=:l/16 and 1/128. In the case of BeClg, however, the proportionality 
fails. 

It will he remarked that the specific gravities of the solutions of beryllium chloride 
for which m = 1/512 and 1/1024 fall below unity, and the values are quite authentic. It 
follows that the displacements of these two solutions must be greater than the sum of 
the displacements of the salt and the water which they respectively contain. A 
similar feature is observed in the saturated solutions of caesium salts, § 127, 
Table III. 

Comparing the values of c?S for concentrations greater than 1/2 gram-molecule, they 
diminish from 0'064018 for MgCL at TO gram-molecule concentration to 0'048345 at 
4*0 gram-molecules, while in the case of CaCla the values are 0"075363 at I'O gram- 
molecule, and 0'058415 at 4*0 gram-molecules concentration. 

The variation in the values of c?S for solutions of CaClg between m = 6*0 and 'm= 6'6 
does not exhibit itself in a regular decrease but an oscillatory one, for the value at m = G'O 
is 0-004541, rising to 0-004810 at w = 6-1, with a fall to 0-004471 at m = 6-2, rising 
slightly again at m = 6'3 to 0-004506, then decreasing to 0-004325 at tn = 6-4 and to 
0*004299 atrn = 6-5, the general tendency being to decrease in value with increasing 
concentration. 

5 101. Values of -^ — and — . — -The features of the displacement of the solutions are 
dm m 

best exhibited by discussing the values of -j— and — . The values of ,— are obtained 

dm m dm 

from columns 7 and 1. 

The solutions of the salts with concentrations less than 1 gram-molecule give values 

dA 
for -^ — which are highest in the case of beryllium chloride, while those of magnesium 

Ctith 

chloride are lowest, those of calcium chloride being intermediate. 



o 



dA 
The value of -^ — for beryllium chloride solution when m=l/2 is 28-66, and this 

decreases to 23-55 when m=l/32, rising to 26-58 at m= 1/128. There are two low 
values, namely, 18-02 and 15-51 at m— 1/256 and 1/512 respectively, with a value of 
27-85 at m= 1/1024. 

In the case of magnesium chloride the value of -j— at m = 1/2 is 2292, and the 
value decreases with succeeding concentrations to the value 17-37 at m= 1/32, which 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 183 



is also the value at w= 1/64. There is a slight rise to the value 17-86 at m= 1/128, 
and a sudden fall to 8*27 at m= 1/256, with an equally sudden rise to 29-80 at m= 1/512. 
The value at m= 1/1024 is 4-10. 

In the case of calcium chloride there is a steady decrease from the value of 25*81 at 



The value 
1/512. The 



w=l/2 for %— to 8-13 where m=l/32, with a rise to 25-14 at m=l/64. 
dm 

decreases to 2 TO 9 at m= 1/256, where there is a sudden fall to 9-11 at m = 

value at w= 1/1024 is 2294. 

The values for v/in for concentrations below m = 1 are on arithmetical grounds more 

dA 
regular in all three cases than the corresponding values for -^. The value for 

beryllium chloride at m=l/2 is 28-04, and this value decreases regularly to 26-37 at 
m= 1/16, where there is a rise to 38-39 at m= 1/128, after which the rate of increase is 
greatly augmented, and reaches a value of 176'13 at m= 1/1024. 

The value oi v/m for magnesium chloride at to = 1/2 is 19-81, decreasing to a value 
of 13-50 at m= 1/128, and rises to 18-74 at m= 1/256, with a sudden fall to 7-68 at 
771=1/512, rising to 11-26 at w= 1/1024. 

Calcium chloride shows the least tendency to sudden variations in the values of 
v/m, since the maximum amplitude is between 22*54 and 14*23 at m=l/2 and 
m= 1/256 respectively. With the exception of the value 22'52 at m = 1/32, there is a 
fairly regular decrease between the two values quoted above, the rate of decrease taking 
place in two phases, the rate being greater between m= 1/32 and 1/256 than between 
m= 1/2 and 1/32. 

There is a rise to a value of 19*35 at m= 1/512, and this decreases to 15*77 at 
w= 1/1024. 

It is seen that the value of v/m at m = 1/2 is highest in the case of beryllium 
chloride and lowest in the case of magnesium chloride, while that of calcium chloride 
more nearly approaches that of magnesium chloride. The rise in the value of v/m for 
calcium chloride, where m=l/2, over that of magnesium chloride is almost exactly 
proportional to the rise in molecular weight. 

§ 102. The values of [dA — v) have been treated in the way fully set forth in 
Section VIII. for the solutions of the salts of the enneads MR and MRO3. It is tliere- 
fore sufficient to give here a table of the values of (dA — v) in the case of solutions of 
wBeCl2, rnMgCl^, and ?nCaCl2 in 1000 grams of water, for which ot<1. 



MR 



BeCl2 
MgCl^ 
CaCl, 



1/2. 



1/4. 



1/8. 



1/16. 



1/32. 



1/64. 



1/128. 



1/256. 



1/512. 



1/1024. 



Values of {dA -v). 



1-557 
1-638 



0-310 


0-098 


0-056 


-0-175 


-0-104 


-0092 


-0-159 


-0-169 


0-614 


0-208 


0-100 


0-026 


0-026 


0-034 


-0-041 


0-043 


0-413 


0-148 


0-074 


-0-137 


0-082 


0-035 


0-027 


-0-020 



-0-145 

-0-007 

0-007 



184 



MR J. Y. BUCHANAN ON THE 



A very remarkable feature exhibited by this table is the pronounced expansion 
which accompanies the dilution of solutions of beryllium chloride for which ■m<l/16. 

§ 103. In columns 5, 9, and 10 of the tables, § 99, we have the data forjudging 
the degree of conformity of the displacements of the solutions of these salts with 
the laws of the two hypotheses, Section VII. They do not agree at all with the 
arithmetic law of the first hypothesis, and their departure from the logarithmic law of 
the second hypothesis is considerable, as may be seen in the columns headed (x — m) 
and (llm—l/x). Thus, in the strong solutions of MgClg and CaCls, taking the ex- 
ponent for m = 1 as unity, we see that the exponents for the solutions of higher 
concentrations increase more rapidly than the corresponding concentration factors m, 
but for both strong and weak solutions x^m. Considering the solutions of BeClg, we 
see that the values of {l/m— '[/a)) change sign for a value of m lying between 1/16 
and 1/32. The behaviour of the solutions of BeClj is, in this respect, quite remarkable. 
If column 6 be referred to, it will be seen that for 



1/128. 
0-2999 



1/256. 

0-2295 



1/512. 
0-1992 



1/1024. 
0-1720 



A - 1000 = 

so that for these very different concentrations the displacements are very nearly the 
same. This depends on the remarkably low specific gravities of these solutions, which 
was commented on in § 100. 

In the following table, which is constructed on the same scheme as Table VIII., § 45, 
the solutions of each salt are taken in successive pairs. The numbers in it represent 

the values -^ — - 



log A 

2 

A^ is taken as unity 
of x should be 2. 



= x, or the exponent of the displacement A^ when the exponent of 
If the solutions conformed to the logarithmic law, the value 



Table giving the Values of ^ '" for the three Salts, Beryllium Chloride, 









Mag 


nesium Chloride, 


and Calcium Chloride. 








m. 


6-0. 


4-0. 


2-0. 


1-0. 


1/2. 


1/4. 


1/8. 


1/16. 


1/32. 


1/64. 


1/128. 


1/256. 


1/512. 


BeClj 

MgClg 
CaClj 


2-154 
2-149 


2-200 
2-253 


2-204 
2-175 


2-145 
2-132 


2-038 
2-127 
2-089 


2025 
2-091 
2-054 


2-049 
2-093 
2-057 


1-807 
2 051 
1-804 


1-795 
2-107 
2-263 


1-692 
2-321 
2-251 


1-307 
1-440 
2-484 


1-152 
4-888 
1-473 


1-158 
1-362 
2-454 



Here again the radical changes which take place in the properties of the solutions 
when the values of m fall below 1/16 are apparent, and particularly so in the case of 
the solutions of BeCL. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 185 

Section XV. — On a Remarkable State of Unrest in a Supersaturated 
Solution of Calcium Chloride before Crystallising. 

§ 104. The primary purpose for which the open hydrometer was designed was to 
investigate the specific gravity and the displacement of solutions having concentra- 
tions in the neighbourhood of that of saturation. In § 90 we have seen the satisfactory 
result of experiments made for this purpose with solutions of chloride of calcium 
containing 6-3 grm.-mols. CaCla per 1000 grams of water. Experiments in the same 
direction were made with solutions of chloride of calcium of still higher concentration. 
A parent solution was made, which, on the basis of published data relative to the 
solubility of the salt, should be supersaturated at 19 "5° C. With it, it was intended to 
produce the solution saturated at this temperature, and to study its specific gravity 
and that of solutions formed by diluting it with small quantities of water. 

As the solution showed no inclination to crystallise, although every opportunity was 
offered to it to do so, it seemed to me to be an example of a supersaturated solution 
peculiarly adapted to closer study. 

Its composition was determined, and it was found to contain 7 '225 grm.-mols. of 
chloride of calcium dissolved in 1 000 grams of water. Its specific gravity was determined 
with the hydrometer exactly as if it had been a non-saturated solution. Two hydro- 
meters were used for this purpose. They are designated A and B respectively. That 
designated A is the hydrometer whose constants have been set out in detail in 
Section XL The hydrometers were made at different dates and on different specifica- 
tions, though possessing the same general characteristics. 

In Table I. are given the constants of both these instruments when loaded so as to 
float with small added weight, (a) in distilled water, and (h) in a supersaturated solution 
of calcium chloride, respectively. The entries opposite " weight of glass " and " weight 
of lead shot " in these tables are the approximate weights in air of these substances, 
which are required for the estimation of the corresponding volumes. The entries 
opposite "weight of the loaded hydrometer" are the exact weights, as in vacuo, of the 
glass -f- lead forming part of each instrument. To each of these weights has to be 
added that of the air contained in the instrument. The external volume of the 
hydrometer is independent of the internal load which it carries. It is entered only in 
(a), line 6. The " internal space occupied by air" is arrived at by subtracting the sum 
of the volumes of glass and of lead from the external volume of each hydrometer 
respectively. The mass of the air which fills this space depends not only on the 
volume of that space, but also on the density of the air which forms the atmosphere of 
the laboratory at the time of the experiment. 



trans. ROY. see. EDIN., VOL. XLIX., PART I. (NO. 1). 24 



186 



MR J. Y. BUCHANAN ON THE 



Table I. 

Constants of Hydrometers A and B when loaded so as to Jloat with small 

added Weight, 

(a) In Distilled Water, and 

(b) In a Supersaturated Solution of Calcium Chloride. 



(a) Distilled Water. 


Hydrometer A, 


Hydrometer B. 


Weight of glass ......... grams 

Volume ,, ......... c.c. 

Weight of lead shot used for internal load .... grams 

Volume „ ,, „ .... c.c. 

Weight of the loaded hydrometer as in vacuo . . . grams 
External volume of hydrometer . . . . . .c.c. 

Internal space occupied by air . . . . . c.c. 

Volume of stem from zero scale division to the top of the stem c.c. 


40-5 

16-2 

95 6 

8-4 

136-1113 

137-09 

112-49 

1-2 


38-11 

15-24 

78-00 

6-87 

115-1026 

118-09 

95-98 

1-0 


(b) SUPERSATDRATED CaLCIUM ChLORIDE SOLUTION. 


Weight of glass ......... grams 

Volume ........... c.c. 

Weight of lead shot used for internal load .... grams 

Volume ,, „ ,, ... c.c. 
Weight of the loaded hydrometer as in vacuo . . . grams 
Internal space occupied by air . . . . . . c.c. 


40-5 

16-2 

148-3 

13-1 

188-8312 
107-79 


38-11 

15-24 
125-4 

11-05 
163-5755 

91-80 



The use of these constants in arriving at the volume of air enclosed in the hydro- 
meter, and in the reduction of the weight of the hydrometer in air to its value in vacuo, 
has been described in Section XI. 

A little consideration and experience enables the experimenter to adjust the internal 
load so that the greatest possible range of specific gravity may be covered without 
altering it. The inferior limit of this range corresponds to a solution of such density 
that the hydrometer floats in it, immersed up to the highest division in the scale, with- 
out the addition of any external weight. The superior limit of the range corresponds 
to a solution of such density that the external weight to be added in order to immerse 
it to the lowest division on the scale begins to endanger the stability of the hydrometer 
as a floating body. 

§ 105. Experiments and Observations with Hydrometers A and B. — The first 
set of determinations of the specific gravity of the supersaturated solution (7 "225 
CaCl2+ 1000 grams of water) was made on 11th May 1910 in one of the smaller rooms 
of the Davy-Faraday Laboratory. The room has a northerly exposure, which is 
essential, and in other respects it is well suited for this class of investigation (§ 24). 

A series of observations had been made with each hydrometer, and further observa- 
tions were proceeding, when it was noticed that discrepancies between successive 



SPECIFIC GRAVITY AND DISPLACEMENT OE SOME SALINE SOLUTIONS. 187 

readings and corresponding ones in the earlier experiments made with the same added 
weights were occurring, and that these were far greater than any which could be attri- 
buted to errors of observation. 

They persisted while four series of observations were made — two sets with each 
hydrometer — and were so great that in the fifth series of observations it was necessary 
to reduce the initial added weight in order that the complete series of observations 
might be made. 

Throughout each series of experiments the temperature of the solution remained 
absolutely constant at 19*50° C. After the removal of hydrometer A from the experi- 
mental solution, on the completion of the fifth series of observations, the solution was 
stirred carefully with the standard thermometer, and its temperature was found to be 
19*50° C, that of the air being 19"30° C. It was not until after these observations 
had been made that a cloudiness indicating the commencement of crystallisation 
appeared in the solution. It increased rapidly, and the temperature rose smartly to 
23 '16° C. and remained constant from 1.10 p.m. to 2.35 p.m. — a period of 85 minutes 
— when the temperature began to fall. 

A careful record of the thermal and other observations throughout the whole 
experiment was kept, and the following is a resume of these data : — 

Weight of solution + cylinder =1270*190 grams. 

Weight of cylinder = 463*580 „ 



Weight of solution taken for observations 

with the hydrometers = 806*610 ,, 

This solution was 7*225 CaClz+lOOO grams of water, and contained 44*48 per 
cent. CaCla- 

§ 106. Thermal Data. — When the hydrometer A had been removed from the 
solution after the fifth series of observations, the time was 1.5 p.m. The solution was 
stirred with the thermometer, gently, and the temperature noted at 1.8 p.m. It was 
at this time that the crystals appeared in the solution, and its temperature rose in less 
than one minute to 23*16° C. and then remained stationary until 2.35 p.m., while that 
of the air in the room varied only between 19*2° and 19*4° C. When the tempera- 
ture of the crystals and the solution had fallen somewhat the cylinder was placed in 
water of 19*3° C. and cooled to 19*5° C, when the mother-liquor was found to have the 
specific gravit)'^ 1*423500 and to contain 42*33 per cent. CaClg. 

§ 107. Rate of Cooling of Original Solution. — The crystals, together with the 
mother-liquor in the cylinder, were then heated to a temperature of 30° C. by placing 
the cylinder in a water-bath of that temperature and keeping it there until the crystals 
were re- dissolved. The system was then allowed to cool in the air, the temperature of 
which remained constant at 19*3° C, and the temperature of the cooling liquid was 
taken at intervals of 30 seconds. 

The series of observations extended over 4 1 minutes, during which the temperature 



188 



MR J. Y. BUCHANAN ON THE 



fell from 23'82° C. to 21*99^ C. and the solution remained liquid to the end. The 
cooling had proceeded for 13 minutes before the temperature fell to 23 16° C, and the 
loss of heat was taking place quite regularly. The following are the temperatures 
observed at each half-minute for two minutes before and two minutes after the 
temperature 23*16° C. was passed : — 

Time in minutes : -2-0 -1-5 - I'O -0-5 O'O +0-5 +1-0 +1-5 +2-0 
Temperature: 23-23° 23*21° 23*19° 23*17° 23*16° 23*14° 23*12° 23*09° 23*07° 

During the four minutes the temperature fell 0*16° C, whence 0*04° C. per minute 
represents the mean rate of fall of temperature when the system has the temperature 
23*16° C. and is cooling in air of constant temperature 19*30° C. 

§ 108. Calculation of Heat liberated during Crystallisation. — The first thermal 
effect observed was when crystallisation began. The temperature of the system rose 
in less than a minute from 19*5° to 23*16°. During this phase the glass cylinder, as 
well as its contents, was warmed 3*66°. The heat liberated in this act depends on 
the weight of the solution, on its specific heat, and on the rise of temperature. In 
determining the thermal exchange which has taken place, we have to take account of 
the capacity for heat, which is generally represented by the " water- value," of the 
cylinder. The numerical data required in this calculation are the following : — 

Weight of CaCla solution ...... 806*61 grams. 

Its specific heat (Regnault) . . . . . . 0*636 

Whence, water- value of solution . .... 513*00 grams. 

Weight of cylinder 463*58 

Specific heat of glass . . . . . . . 0*2 

Whence, water-value of cylinder .... 92*72 grams. 

Rise of temperature ....... 3*66° C. 

Whence, heat liberated in first act (51300 -I- 92*72)3*66 = 2217 gr.° C. 

After the first minute, when the temperature had become constant at 23*16°, the 
rate of liberation of heat was exactly equal to its rate of dissipation, which we have 
found to be represented by a fall of temperature of 0*04° per minute. This state 
was maintained for 85 minutes, which requires a liberation of heat, in the second act, of 

85 X 605*72 x 0*04 = 2059 gr.° C. 

Adding the 2217 gram-degrees liberated in the first act, we find the total heat evolved 
during the interval of 85 minutes to be 

2059 -f 2217 = 4276 gr.° C. 

§ 109. In order to verify the state of unrest above described, the experiment was 
repeated with a 7*196 CaClg solution (§ 113), and after crystallisation was completed, 
the crystals were removed and freed as far as possible from adherent mother-liquor, 
and their composition ascertained by estimation of the chlorine contained in a weighed 
quantity. The results of duplicate determinations gave the composition of the crystals 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 189 

SO obtained as CaClaG-SHaO and CaCl^e-dHgO respectively. It is obvious, therefore, that 
the crystals which were deposited had the composition CaClaGfTaO, and that the excess 
of water indicated by the analyses was due to some adherent mother-liquor, from which 
it is almost impossible to free the crystals. 

That the crystals deposited in the first experiment had the composition CaCLGHaO 
is confirmed by the thermal data already set forth. 

The weight of original calcium chloride solution which crystallised in the cylinder 
was 806'61 grams. It contained 44'48 per cent. CaClz. 

The mean of the first series of specific gravities— 1 '4 460 19 — is taken as the specific 
gravity of this solution when at rest. 

The concentration of the solution is therefore : — I 

Per Cent. By Weight. 

CaClj . . 44-48 359 '0 grams. 

Water . . 55-52 447 6 „ 



Total . . 100-00 806-6 „ 

After the crystallisation was ended, and with the solution at a temperature of 
I9'5° C, one determination of the specific gravity of the mother-liquor gave the 
result 1-423500. 

The concentration of the mother-liquor determined by analysis was 42'33 per cent., 
equivalent to 734"0 grams, or 6'613 gram-molecules CaCla per 1000 grams water. 

The crystals (CaClgeHgO) contain 50"685 per cent. CaCla- 

The cooling observations showed that the heat evolved in the act of crystallising 
was 4276 gr.° C. 

According to Thomsen, the heat of solution of CaClaeHgO is - 4340*0 gr.° C. ; there- 
fore on thermal evidence alone 215'5 grams, or 0'984 CaClzBHaO, has separated out. 
But, on the basis of the analytical estimations made on the supersaturated solution 
and the mother-liquor, we find that 210*3 grams of crystals separated out of 80661 grams 
of solution, or 096 gram-molecule CaClaeHaO. The agreement of these two computed 
values is excellent. 

We accept then as the quantities of crystals and mother-liquor 210*3 grams and 
596 '3 grams respectively. 

§ 110. The nature of the experiments having been indicated, and the general 
character of the thermal change and the alterations in specific gravity mentioned, the 
following table, IIa., gives a complete account of the individual observations of specific 
gravity made, together with the corresponding displacements calculated from them. 

Table IIb, gives a similarly complete account of the individual observations of 
specific gravity in five series made in the solution 6*3 CaClg-f- 1000 grams of water at 
19*5° C, with the corresponding displacements calculated therefrom. The solution of 
calcium chloride saturated at 19*5° C. is 6613 CaClg-t-lOOO grams of water; therefore 
the 6*3 CaClg solution, though of high concentration, is sufficiently removed from 
saturation to exhibit the tranquillity of a dilute solution. 



190 



MR J. Y. BUCHANAN ON THE 



Table IIa. 

Table of Observations made with Hydrometers A and B ivhen jioating in the 
Supersaturated Solution 7"225 CaCl^-h 1000 grams of Water at 19"5° C. 
before Crystallisation. 





Hydrometer A. 








Hydrometer B. 




Series 1 (10.45 a.m.-11.5 a.m. 


)• 




Series 2 (11. 3C 


a.m. -11. 45 a.m.). 


Added 

Weight in 

Grams. 


Scale 
Reading in 
Millimetres. 


Specific Gravity 

calculated 

from Single 

Observations. 


Displace- 
ment. 


Added 

Weight in 

Grams. 


Scale 
Reading in 
Millimetres. 


Specific Gravity 

calculated 

from Single 

Observations. 


Displacement. 


8-65 


13-7 


1-445961 


1245-71 


6-1 


7^8 


1-445905 


1245-76 


8-75 


20-9 


5972 


-70 




6-2 


16^9 


5826 


•83 


8-85 


28-1 


6034 


-64 




6-3 


25^3 


5837 


•82 


8-95 


35-6 


6044 


-63 




6-4 


34^0 


5828 


•83 


9-05 


42-8 


6033 


-64 




6-5 


42-1 


5880 


•78 


9-15 


50-4 


6015 


-66 




6-6 


50-7 


5895 


•77 


9-25 


57-8 


6039 


■64 




6-7 


58-3 


5962 


•72 


9-35 


65-0 


6049 


-63 




6-8 


67^1 


5948 


•71 


9 45 


72-4 


6037 


•64 




6-9 


75^8 


5936 


•64 


9-55 


79-9 


6001 


■68 




7-0 


83-6 


5998 


•67 


9-65 


87-1 


6028 


-65 




7i 


91-3 


6076 


•61 


Hydrometer A. 


Hydrometer B. 


Hydrometer A. 


iSeries 5 


1(12.10 p. m.-12. 25 p.m.). 


Series 4 (12.35 p.m.-12.49 p.m.). 


Series 5 (12.55 p.m. -1.5 p.m.). 




Scale 


Specific 






Scalp 


Specific 






Scale 


Specific 




Added 
Weight 


Reading 


Gravity 
calculated 


Displace- 


Added 
Weight 


Reading 


Gravity 
calculated 


Displace- 


Added 
Weight 


Reading 


Gravity 
calculated 


Displace- 


m 


Milli- 
metres. 


from Single 


ment. 


in 


Milli- 
metres. 


from Single 


ment. 


in 


Milli- 
metres. 


from Single 


ment. 


Grams. 


Observa- 
tions. 




Grams. 


Observa- 
tions. 




Grams. 


Observa- 
tions. 


1245-92 


8-65 


16-0 


1-44.5741 


1245-90 


61 


7-8 


1-445905 


1245-76 


8-6 


12-5 


1-445719 


8-75 


23-9 


5780 


-87 


6-2 


16-9 


5826 


•83 


8^7 


201 


5703 


■93 


8-85 


31-6 


5686 


•95 


6-3 


25-3 


5837 


•82 


8-8 


27-9 


5681 


■95 


8-95 


39-1 


5663 


-97 


64 


34-3 


5796 


-86 


8-9 


36-9 


5525 


1246^09 


9-05 


46-5 


5711 


-93 


65 


42-7 


5820 


-84 


90 


45-1 


5448 


•16 


9-15 


53-6 


5831 


-82 


6^G 


51-8 


5794 


-88 


91 


53-9 


5313 


•27 


9-25 


61-0 


5717 


•92 


6-7 


60-0 


5852 


•81 


9-2 


62-9 


5165 


■40 


9-35 


68-1 


5732 


-91 


6-8 


68-1 


5855 


-80 


9-3 


70-5 


5133 


■43 


9-45 


75-3 


5768 


-88 


6-9 


77-1 


5805 


■79 


9-4 


78-2 


5096 


•46 


9-55 


82-6 


.5796 


-85 


7-0 


85-2 


5838 


-82 


9-5 


86-8 


4959 


•58 


9-6.5 


89-9 


5824 


-83 


71 


94-5 


5759 


•89 


9-6 


94-7 

1 


4886 


•64 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 191 



Table 11b. 
Table of Observations made with Hydrometers A and B when floating in the 
Solution 6-3 CaC4+ 1000 grams of Water at 19-5° C, for Comparison with 
the Results given in Table II a. 



Hydrometer A. 






Hydrometer B. 


Series 1. 




Series 2. 


Added 


Scale 


Specific Gravity 

calculated 

from Single 

Observations. 




Added 


1 


Scale 


Specific Gravity 

calculated 

from Single 

Observations. 






Weight in 
Grams. 


Reading in 
Millimetres. 


Displacement, 




Weight in 
Grams. 


Reading in 
Millimetres. 


Displacement. 


4-0 


13-9 


1-409760 


1204-94 


2-05 




7^1 


1-409740 


1204-96 


4-1 


•21-8 


735 


•97 




215 




15^9 


732 




-96 


4-2 


29-6 


731 


-96 




2^25 




24^5 


744 




-95 


4-3 


37-2 


730 


•96 




235 




335 


725 




-97 


4-4 


44-6 


744 


•95 




2^45 




42^8 


683 


120500 


4-5 


52-2 


749 


-95 




2^55 




50-8 


751 


1204-95 


4-6 


59-8 


762 


-94 




2^65 




600 


723 




•97 


4-7 


671 


776 


-92 




2-75 




679 


725 




-97 


4-8 


74-7 


764 


-93 




2-85 




77^8 


694 




-99 


4-9 


82-2 


762 


-93 




2-95 




86^0 


740 




-96 


5-0 


89-7 


761 


-94 




3-05 




94^8 


739 




•96 


Mean . . 1-409752 


Mean 


L 


• 


1-409727 




Probable error of mean expressed in \ _|_ „ _o.ii 
units of the sixth decimal place / ~ ' ^ 


4^46 


Hydrometer A. 


Hydrometer B, 


Hydrometer A. 


Series 3. 


Series 4. 


Series 5. 




Soalp 


Specific 






Slnnlp 


Specific 








Scale 


Specific 




Added 
Weight 


Reading 


Gravity 
calculated 


Displace- 


Added 
Weight 


Reading 


Gravity 
calculated 


Disj 


lace- 


Added 
Weight 


Reading 


Gravity 
calculated 


Displace- 


iu 
Grams. 


Milli- 
metres. 


from Single 
Observa- 
tions. 


ment. 


in 
Grams. 


Milli- 
metres 


from Single 
Observa- 
tions. 


me 


nt. 


in 
Grams. 


Milli- 
metres. 


from Single 
Observa- 
tions. 


ment. 


4-0 


14-1 


1-409741 


1204-95 


2-05 


7^0 


r409750 


120' 


^•95 


4-0 


140 


r409750 


1204-95 


4-1 


21-7 


745 


-95 


215 


15^8 


742 




•95 


4-1 


21-5 


764 


•94 


4-2 


29-7 


722 


-97 


2-25 


24^7 


724 




•97 


4^2 


29-5 


741 


-95 


-I-3 


37-5 


730 


-96 


2-35 


33^8 


696 




•99 


43 


37-4 


711 


-98 


4-4 


44-9 


722 


-97 


2-45 


42^0 


760 




•94 


4-4 


44-8 


731 


-96 


4-5 


52-1 


765 


-94 


2-55 


50^9 


742 




•95 


4^5 


52-1 


765 


■94 


4-6 


59-8 


768 


-93 


2-65 


59^0 


818 




•89 


4^6 


610 


741 


•95 


4-7 


67-2 


772 


-93 


2-75 


68^1 


790 




•92 


4^7 


67-5 


744 


•95 


4-8 


74-9 


750 


-95 


2-85 


76^9 


782 




•92 


4^8 


74^8 


760 


•94 


4-9 


82 4 


748 


-95 


2-95 


86^0 


740 




•95 


4-9 


82-6 


730 


•96 


5-0 


89-9 


749 


-95 


3^05 


94^8 


739 




•96 


5-0 


89^7 


767 


-93 


Mean . . 1-409746 


Mean . . 1 409753 


Mean 


. r409746 




Probable error of mean \ 












expressed in units of the |- ± r-g = 3-49 


6-35 






3-6 




sixth decimal place j 













192 



MR J. Y. BUCHANAN ON THE 



Table lie. 

Table of Observations made with Hydrometers A and B in the Supersaturated 
Solutions, 7-196 CaCl^-\- 1000 grams of Water, at 19-50° C. 



Index of 


Hydro- 
meter 
used. 


Time Interval in 










Series of 


Minutes of Successive 


Added Weight 


Reading in 


Corresponding 


Correspoiidiiic/ 


Observa- 


Observations from 


in Grams. 


Millimetres. 


Sjiecific Gravity. 


Displacement. 


tions. 


first Observation. 














1-4 


8-65 


13-4 


1-443843 


1245-81 






2-7 


8-75 


20-5 


876 


•78 






4-1 


8-85 


28-0 


884 


•78 






5-5 


8-95 


35-6 


868 


•79 






6-8 


9-05 


43-1 


859 


•80 


1st 


A 


8-2 


9-15 


50-9 


824 


•83 






9-6 


9-25 


58-2 


851 


•80 






10-9 


9-35 


65-8 


825 


•83 






12-3 


9-45 


73-0 


831 


-82 






13-6 


9-55 


80-3 


832 


-82 






15-0 


9-65 


87-5 


843 


•81 


Time interval of 8 minutes between experiments. 






24-5 


6-1 


10-0 


1-443941 


1245-73 






25-9 


6-2 


18-8 


911 


-75 






27-4 


6-3 


27-3 


912 


-75 






28-8 


6-4 


36-7 


839 


•82 






30-3 


6-5 


45-0 


867 


■79 


2nd 


B 


31-7 


6-6 


54-2 


806 


-84 






33-2 


6-7 


62-6 


830 


-82 






34-6 


6-8 


711 


830 


•82 






36-1 


6-9 


79-9 


807 


■84 






37-5 


7-0 


88-0 


844 


-81 






39-0 


7-1 


96-6 


838 


•82 


Time interval of 8 minutes between experiments. 






48-5 


8-65 


13-9 


1-443793 


1245^85 






501 


8-75 


21-3 


799 


■85 






51-6 


8-85 


28-2 


865 


■79 






53-2 


8-95 


36-6 


769 


•88 






54-7 


9-05 


43-8 


800 


-85 


Srd 


A 


56-3 


9-15 


51-5 


767 


-88 






57-8 


9-25 


58-3 


841 


-81 






59-4 


9-35 


• 66-2 


784 


-86 






60-9 


9-45 


73-8 


752 


-89 






62-5 


9-.55 


81-0 


762 


-88 






64-0 


9-65 


88-7 


724 


•92 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 193 



Table III. 

Giving Details of the Order of Succession of the Experiments, of their Duration, 
and of the Temperatures of the Solutions and the Air respectively during 
the Experiments made on the Solution, 7 '225 CaCl^^- 1000 grams of Water. 



Number of 
Experiment. 

(1) 


Time of Com- 
mencement of 
Experiment. 

(2) 


Time of Com- 
pletion of 
Experiment. 

(3) 


Duration of 
Experiment. 

(4) 


Time Interval 

between 

Successive 

Experiments. 

(5) 


Initial and 
Final Solution 
Temperatures. 

(6) 


Initial and 

Final Air 

Temperatures. 

(7) 


Hydrometer. 
(8) 


1 


10.45 a.m. 


11.5 a.m. 


20 minutes 


25 minutes 


19-5° C. 
19-5° C. 


19-3° C. 

19-rc. 


A 


2 


11.30 a.m. 


11.45 a.m. 


15 minutes 


25 minutes 


19-5° C. 
19-5° C. 


19-3° C. 
19-3° C. 


B 


3 


12.10 p.m. 


12.25 p.m. 


15 minutes 


10 minutes 


19-5° C. 
19-5° C. 


19-3° C. 
19-4° C. 


A 


4 


12.35 p.m. 


12.49 p.m. 


14 minutes 


6 minutes 


19'5°C. 
19-5° C. 


19-4° C. 

19-3° C. 


B 


5 


12.55 p.m. 


1.5 p.m. 


10 minutes 




19-5° C. 
19-5° C. 


19-3° C. 
19-35° C. 


A 



§ 111. Comparison of Results obtained with Hydrometers A and B when floating 
in the Supersaturated Solution of Calcium Chloride with those obtained when the Hydro- 
meters are floating in a Solution containing 6 "3 gram-molecules of Calcium Chloride in 
1000 grams of Water. — We will first draw attention to Table III., which gives the dura- 
tion of each experiment, the initial and final solution temperatures, and the air tempera- 
tures before and after each experiment, in connection with the observations and results 
recorded in Table IIa. Table III. also affords a fair criterion of the usual duration 
of these experiments, and it gives suitable relief to the constancy of the temperature, 
both of the experimental liquid and of the atmosphere of the laboratory, which can, 
and must be secured, if the full precision of which the hydrometric method is 
susceptible is to be achieved. It will be observed that the temperature of the air 
was generally 0-2" C. lower than that of the experimental liquid, and that, in these 
conditions, the temperature of the experimental liquid remained perfectly constant 
during the time of the experiment. The absolute degree of constancy covered by this 
statement depends on the specification of the thermometer used. It was a standard 
instrument, divided, on the stem, into tenths of a Centigrade degree, and the length 

TRANS. fiOY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 25 



194 



MR J. y. BUCHANAN ON THE 



of a whole degree was 12 millimetres. On such an instrument variations of one- 
hundredth of a degree are easily appreciated by the practised eye. It must, however, 
never be forgotten that ivliat is directly observed is, at the best, the most probable 
value of the temperature of the bulb of the thermometer. The legitimacy of the 
conclusion that this is the temperature of the medium in ivhich the thermometer is 
immersed depends on the expertness and experience of the experimenter. The use 
of an instrument of the degree of delicacy above specified is justified only when all 
the precautions have been taken which are required in order to justify the experi- 
menter in concluding that he has the temperature of the system under such control 
that its uncertainty is not greater than ±0*005° C. Nothing short of first-rate 
work secures this. 

In the work which Mr S. M. Bos worth has been doing under my direction, upwards 
of 3000 hydrometer observations have been made during the last twelve months, and 
the temperature conditions have been controlled with such skill that only three of the 
series showed a sensible variation of the temperature of the solution from the standard 
temperature during the time the experiment lasted. Their results were rejected, not 
because they were not very good, but because in this respect the others were perfect. 

We will now proceed to a comparison of the figures in Tables IIa. and IIb. 

§ 112. It would be useless to take the means of each series of specific gravity 
results in Table IIa. and compare them with the mean results given in Table IIb. 
But these numbers show the progressive character of the alteration of the specific 
gravity in each consecutive series, as well as the considerable diff"erences of these 
numbers inter se, when contrasted with the agreement which holds among the mean 
results set out in the other table, IIb. 

The mean results are given in the following table : — 

Table IV. 
Giving the Mean Specific Gravities calculated from the Series in Table IIa. 



Series. 


Hydrometer. 


Mean Specific Gravity. 


1 
2 
3 
4 
5 


A 
B 
A 
B 

A 


1-446019 
1-445917 
1-445750 
1-445826 
1-445330 



Taking the five mean results given above, we find that the maximum amplitude of 
variation is 689 units in the sixth decimal place. If we turn to Table IIb., we find 
the maximum amplitude of variation to be only 26 such units. 

Although the consideration of the mean specific gravities shows clearly that there 
is an unstable condition in the supersaturated solution as contrasted with the stable 
condition of the 6 "3 CaClg solution, this is made more evident if we compare, as in 



1 


2 


3 


4 


5 


A 


B 


A 


B 


A 


88 


250 


168 


146 


833 


46 


68 


50 


66 


56 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 195 

the following table, the greatest amplitude of the variation in each series given in 
Table 1 1 a. with those occurring in series made with the same hydrometer, as given in 
the other table, IIb. 

Series ...... 

Hydrometer ..... 

Amplitudes of Variation < 

^ I Table 11b. 

In series 4 of Table IIb. there are two extreme values of the specific gravity, 
namely, 1*409696 and l"409818, giving an amplitude of 122, but they are quite ex- 
ceptional, and if they are omitted the greatest amplitude in the series is 66, so that 
the mean maximum amplitude in the series made with hydrometer B is &7 , while that 
in the series made with hydrometer A is 50. 

On examining the figures for the supersaturated solution, their irregularity when 
compared with those of the G'3 CaClg solution is at once apparent; moreover, the 
irregularity increases very rapidly in each consecutive series, so that the amplitude of 
variation is in the first series 88, rising to 250 in the second, then falling to 146, and 
reaching 833 in the last series, immediately after which crystallisation commenced. 
I The rate of increase is not regular, for in the fourth series the variation amounts to 

I 146, which seems to show that the solution was for the time being in a state of com- 
parative calm. 

This table, therefore, serves the useful purpose of indicating the fluctuating 
j character of the alteration of the specific gravity, while Table IV. shows that in 
' spite of these fluctuations there is a definite decrease in specific gravity from the first 
to the fifth series. 

§ 113. Displacement. — The displacement of the 6 "3 gram - molecule solution 
throughout the series is for all practical purposes constant, as might be expected, 
since the solution is quite stable and no disturbing influences, such as the imminence 
of crystallisation, are present. 

But the displacement in the case of the supersaturated solution is subject to variations 
corresponding to those of the specific gravity. These afford evidence of considerable 
and, to some extent, spasmodic acts of expansion and contraction, unaccompanied hy 
any change of temperature of the solution or of the external pressure to which it ivas 
subjected. These spasmodic changes of volume exhibit a veritable species of labour, 
going on in the solution in its efforts to become a mother-liquor. In this it is finally 
successful, but not before it has succeeded in forcing the door which confned its store 
i of heat. The birth of the crystal ivas synchronous with, and dependent on, the libera- 
tion of heat. 

If we consider the mean specific gravities of the five series given in Table IIa., we 
find a progressive decrease in them from the first to the fifth, with an interruption in 
the fourth, where a slight increase is observable over that of No. 3. This resultant 
decrease of the mean specific gravity of the solution is accompanied by a series of 



196 



MR J. Y. BUCHANAN ON THE 



fluctuations in the results of the single observations of each series which indicates a 
condition of unrest in the solution, which is most accentuated in the fifth series of 
observations, during which the specific gravity fell from 1 "445719 to 1 "444886. The 
most remarkable feature of these changes is that they occurred without being ac- 
companied by any change in the temperature of the solution. Confirmation of the 
occurrence of this remarkable condition of unrest was furnished by a repetition of the 
experiment, made with a solution containing 7 "196 gram-molecules CaClg per 1000 
grams HgO, recorded in Table lie, and in it similar fluctuations of density, although 
not quite so pronounced, occurred as the forerunner of crystallisation. When the 
temperature of the solution had been observed after completion of the third series, the 
side of the cylinder was accidentally rubbed by the thermometer and crystallisation took 
place; but the behaviour of this solution and that of the 7"225 CaClz solution, in the 
case of the first three series, exhibit similar features of unrest. 

§ 114. It is interesting to inquire into the nature of the changes of displacement 
which have occurred in the transition of the solution from a condition of supersaturation 
to that of a mixture of saturated solution and crystals at the same temperature, 19 "5° C. 
They are clearly shown in the accompanying table. 

We commence with 806"61 grams of a solution having a specific gravity 1-444886, 
the displacement of this weight of solution being therefore .558 "25 grams, and this re- 
solves itself into 596"3 grams of mother-liquor with a specific gravity of 1 "423500, giving 
a displacement of 418'89 grams, and 210"3 grams of CaCl26H20 crystals with a specific 
gravity of 1*654 (Landolt's Tables), giving a displacement of 127*15 grams, from which 
we see that the displacement of the mixture of crystals and mother-liquor is 12*21 grams, 
or 2 "2 per cent less than that of the original volume of supersaturated solution. 

The following table shows this clearly : — 





Weight in 
Grams. 


Displacement in 
Grams of Water. 


Volume in c.c. 
at 19-5° C. 


Percentage 
Displacement. 


Original solution 

Mother-liquor . 
CaClgeHgO 

Total 

Diflference . 


806-6 


558-25 


558-75 


10000 


596-3 
210-3 


418-89 
127-15 


419-48 
127-36 


75-03 
22-83 


806-6 


546-04 


546-84 


97-86 


... 


12-21 


11-91 


2-14 



§ 11 5. When we compare the behaviour of the chlorides of magnesium and calcium 
in supersaturated solution, it seems strange that the salt which has the greater amount 
of heat to lose by crystallising should be the more difiicult to bring to crystallisation. 
The difficulty, however, lies only in the starting of crystallisation ; there is none in its 
continuation. To start crystallisation in any solution, no matter how supersaturated 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 197 

it may be, is an operation which has some of the elements of an act of creation — some- 
thing appears where there was nothing of the kind before. The actions and reactions 
which take place before the first element of crystal appears are withdrawn from our 
sight, but their existence has been revealed by the hydrometer, which faithfully reports 
to him who can use it the dilatations and contractions which precede the crystal's birth. 

They are illustrated in the accompanying diagram. In it the ordinates represent 
displacements of the solution and the abscissae intervals of time, dating from the first 
observation of the first series. The lowermost curve. No. 1, represents graphically the 
data of displacement relating to the supersaturated solution 7*225 CaCla+lOOO grams 
of water given in Table IIa. The uppermost curve, No. 2, represents those relating to 
the non-saturated solution 6*3 CaCl2+ 1000 grams of water given in Table IIb. ; and the 
intermediate curve. No. 3, represents the data relating to the supersaturated solution 
7"196 CaCla+lOOO grams of water given in Table lie. 

The displacements are plotted in the order in which they were observed, and the 
series follow each other in chronological order. The letters A, B designate the hydro- 
meter which was used for the series represented under each respectively. The time 
intervals between successive series of observations are included in the diagram and are 
traced by dotted lines. The interval of time which separated two consecutive series 
of observations was on an average seventeen minutes, and the duration of each series of 
experiments was about fifteen minutes. The diagram shows at a glance the contrast 
between the tranquillity of the 6 '3 CaClg solution and the unrest indicated by the 
curves for the two supersaturated solutions. 

The curve No. 2 for the 6*3 CaClg solution pursues an even course, the displacements 
oscillating between the extremes 1205'00 and 1204'89. Otherwise the curve differs 
little from a straight line, and there is perfect agreement between the last result in one 
series and the first in the succeeding series. This is shown by the horizontality of the 
four dotted lines connecting the successive serial curves. 

Curve No. 1 for the 7*225 CaClg supersaturated solution is in striking contrast with 
No. 2. There is little agreement between the displacements in any of the corresponding 
series, and the oscillations of the serial curves are very marked, culminating in the 
continuous expansion shown in series 5, after which crystallisation took place. This is 
also well shown by the difference between the last displacement of one series and the 
first displacement of the following series. It is evident that the state of unrest con- 
tinued when the solution was left to itself in the cylinder. The slight contraction shown 
in passing from the third to the fourth series indicates an effort on the part of the 
solution to regain a more stable condition. It is, however, clear that the state of unrest 
continued during the whole of the 140 minutes represented by the line of abscissae in 
the diagram, and it suggests the possibility, and indeed the probability, that the super- 
saturated solution, even when confined in a closed vessel, may never be at rest. 

§ 116. If we consider attentively what took place before the supersaturated solution 
of calcium chloride was brought to shed its salt as crystals, it is seen that it differs very 



198 



MK J. Y. BUCHANAN ON THE 







oc 




>o 


■" 


't 


cs 


o 




a 


>o 




oo 








CM 




— 








CS 


to 

s 


to 
Csl 


o 


















^_^ 












— bl 

(- 


D 
















' ~ 


~ _^ 










D 


5to 
















< 




"^ 








Z 

o - 

rt2 




























o 

o 




















^ 


V 


> 




o 
o 


»<5 

o 




o 




.X 


























/ 




— 
























f 














/ 












i 








o 

o 








> 












1 






o 
o 






< 


I 










-"C 








o 








> 










<^ 








o 








1 










s 








o 

00 








1 












\ 






o 

00 








\ 






O- (v. 






\ 






o 






CD 


/ 


• 




to lo 








\ 

\ 
\ 




o 








1 

t 




















\ 




o 








1 
1 








> 


V 










\ 




o 

•o 








I 










> 










»< 










< 


) 








< 


c 










/^ 






o 








r 










> 






^ 


/ 




O 

to 




d 




\ 






o 

X" 




r 

1 






o 

x" 








o 

O 

to 




X 














1 






§ 




I 




o 

o 

to 




1 

o 
o 


CD 








4 
^- 
00 

o 
o 
o 




i 






O 

o 
o 


\ 






o 










+ 




< 


*»»«H 




+ 




\ 




o 

CM 




+ 




1 






u 

•i 

o 






/ 




ffs* 

o 




I 




o 




o 
to 




1 
\ 






2: 




i 


/ 




CM 




<) 




o 




>o 




\ 






t>^ 




i 






P«N 






o 




(S 




/ 






rO 




< 






— 




\ 








o 

z 


< 


/ 






01 




< 


^ 




Oi 




/ 






o 








S 










y 








/ 




o 






— 


O 


o. 


oC 




o 


o 


00 


tv 






00 


•o 


^ 




DO 


o 


lO 

1 


O 


o 






lO 




"3 






lO 

^ 

C; 


"3 


I 1 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 199 

little from what takes place when its non-saturated solution is brought to shed part of 
its water as crystals of ice. In each case considerable depression of the temperature 
of the solution below that of crystallisation is required before crystallisation sets in. 
When it does set in, an immediate rise of temperature takes place in the solution, and 
it stops only when the ordinary temperature of crystallisation has been reached. The 
same sequence of events is observed when water, containing no salt, is brought to 
crystallisation ; it also rises to its ordinary temperature of crystallisation. Unless these 
experiments are made in conditions which are unusual in chemical laboratories, none of 
these liquids begins to crystallise immediately when its temperature has been lowered 
exactly to the crystallising point. 

The extent to which any particular solution can be cooled below its crystallising 
temperature, and the amount of mechanical disturbance which it can withstand when in 
this condition, vary with the nature of the solution. We have seen that the resistance 
so offered by a supersaturated solution of calcium chloride is considerable. But even 
that offered by pure water to the starting of crystallisation is greater than is generally 
believed to be the case. 

When the mass of water used is small, and the capacity for heat of the vessel which 
contains it is large, it is possible, with care, to reduce the temperature of the system so 
far that, when crystallisation takes place, the whole of the water is transformed into 
ice and its temperature does not rise so high as 0° C. An instance of this is given in 
the following passage quoted from a lecture on " Ice and its Natural History " which I 
delivered before the Royal Institution on 8th May 1908 : — * 

" Evidence of the uncertainty which exists regarding the temperature at which ice 
begins to form in water, when it is cooled in contact only with a solid other than ice, 
is furnished by the wet-bulb thermometer when it is being prepared for use at tem- 
peratures below 0° C, by freezing on it the quantity of water which is supported, against 
gravity, by the perfectly clean bulb. When this is rotated in air of — 10° to — 20° C, 
ice never begins to form until the temperature of the bulb of the thermometer has fallen 
to —2° or —3° C, and rarely before it has fallen to -4° C. In many cases I have 
observed it fall to temperatures as low as —8° or —9° C. ; and in such cases, when 
freezing begins, the whole of the water is frozen without its being able, by the liberation 
of latent heat alone, to raise the temperatui-e of the bulb of the thermometer to 0°C." 

Whether, when in this unstable state, it would stand the mechanical disturbance 
which is resisted by a supersaturated solution of calcium chloride, can only be determined 
by experiment. This I have not as yet attempted. All that is required is to set about 
determining the specific gravity of pure water hydrometrically in a laboratory having 
the constant temperature — 4° or — 5° C, and using at least equal precautions with those 
observed in the case of supersaturated solutions at ordinary room temperatures. 

If water, in these conditions, is sufficiently unsensitive to mechanical disturbance, it 
will undoubtedly do its part in manifesting its unrest ; it will then be the part of the 

* Proceedings of the Royal Institution of Great Britain, 1909, xix., Part I. p. 248, 



200 MR J. Y. BUCHANAN ON THE 

experimenter to receive the message, and if he succeeds he will have done a very fine 
piece of work. 

§ 117. In considering the dilatation of the 7'225 CaCla solution before crystallisation, 
we may pass over the first four series, although the oscillations of displacement 
which they exhibit would be remarkable enough if they stood alone, and confine our 
attention to the expansion in the fifth series which is continuous from the first to the 
eleventh observation. The displacement at the first observation was 1245'92, and at 
the eleventh 1246 "64, corresponding to an increase of 072 gram in ten minutes. But 
the absolute minimum displacement observed was 1245*61 in the second series, so that 
the extreme amplitude of expansion was 103 gram. While these changes of displace- 
ment were going on, the liquid was perfectly homogeneous and its temperature was 
absolutely constant. Therefore the dilatations were not of thermal but of mechanical 
origin. We can apply no mechanical power which would produce such a stretching 
eflfect on a liquid, but we can easily arrive at the mechanical power which could effectually 
counteract it. 

Drecker * gives 0*0000217 as the coefficient of compressibility of a 40 '9 per cent, 
solution of CaCl2, and we may take this as the compressibility of our 7*225 solution, 
although it would be rather less. On this basis we obtain 38 atmospheres as the 
pressure required to reduce the volume of the solution 7*225 CaCl2+ 1000 grams of water 
from 1246*64 to 1245*61 cubic centimetres, and we conclude that, if we could place the 
solution in conditions such that its internal pressure should be increased hy 38 atmo- 
spheres, the extreme dilatation obsemed wo/dd he mechanically provided for. These 
isothermal oscillations cease immediately when the first element of crystal appears in 
the solution and aff"ords an outlet for its latent heat, after which crystallisation proceeds 
in perfect tranquillity at a rate proportional to that at which heat is removed from the 
solution. It is stopped if heat is supplied at this rate to the solution from without. 

§ 118. There is a remarkable resemblance between the state of unrest preceding 
the crystallisation of a supersaturated solution and that preceding the liquefaction of 
a gas, under a pressure not inferior to its critical pressure, when its temperature is 
reduced slightly below its critical temperature. 

Andrews, in reporting his discovery of the critical state of liquids and gases, in- 
cidentally describes this state of unrest as follows : — " On practically liquefying carbonic 
acid by pressure alone, and gradually raising at the same time the temperature to 
88° Fahr. (31*1° C), the surface of demarcation between the liquid and gas became 
fainter, lost its curvature, and at last disappeared. The space was then occupied by a 
homogeneous fluid which exhibited, when the pressure was suddenly diminished or the 
temperature slightly lowered, a peculiar appearance of moving or flickering striae through- 
out its entire mass. At temperatures above 88° Fahr. no apparent liquefaction of 
CO2 or separation into two distinct forms of matter could be effected, even when a 
pressure of 300 or 400 atmospheres was applied " {Phil. Trans. (1869) vol. clix. p. 575). 

* Wied. Ann., 1888, vol. xxxiv. p. 955. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 201 

When a gas such as carbonic acid, under a pressure which is not inferior to its 
critical pressure, is confined in a tube as in Andrews' experiment and has a temperature 
ever so little higher than its critical temperature, it fills the tube as a homogeneous 
fluid which no pressure, however high, can liquefy. If, by removal of heat or by sudden 
relief of pressure, its temperature is reduced ever so little below its critical temperature, 
the homogeneous fluid begins to exhibit the peculiar appearance of moving or flickering 
striae throughout its entire mass, as described by Andrews. These moving or flickering 
striae indicate oscillations of density accompanying the efi"ort on the part of the homo- 
geneous fluid to shed a portion of its mass in the liquid state before there is a liquid 
nucleus for it to condense on and to aff"ord the first outlet to the latent heat, the escape 
of which is an essential condition of liquefaction. 

§ 119. We do not know the temperature at which the dry gas can condense on the 
dry walls of the envelope, but there can be little doubt that it is lower than that at 
which it condenses on its own liquid. 

It is only in the conditions of Andrews' experiment that we can witness a substance 
persisting in the gaseous state under a greater than the critical pressure and having a 
temperature lower than the critical temperature, because it is only when the gas and 
the envelope which contains it have been maintained at a temperature higher than 
the critical temperature of the gas that the inner walls of the envelope have a chance 
of being perfectly dry. By " perfectly dry " I mean free from every and any trace 
whatever of the liquid substance. If the cooling process as specified above be then 
carried out, the temperature may be reduced slightly below the critical temperature, 
and yet the substance may persist in the gaseous state because there is none of itself 
in the liquid state for it to begin to condense on. 

I have defined * the boiling point of a substance, under a particular pressure, to be 
the temperature at which it as a liquid evaporates into itself as a gas, arid as a gas or 
vapour condenses on itself as a liquid. When the gas condenses on any other substance, 
or in a space filled only by itself, the temperature at which liquefaction commences is 
uncertain. In the moment, however, that the first, even the minutest trace of liquid 
appears, whether in the gas or on the walls, the temperature of condensation is defined, 
because the gas is then condensing on itself as a liquid. How, in such an experiment, 
the first element of liquid appears, we do not know. We say, it is by accident ; but 
we may with equal right say, it is by an act of creation — because in the process some- 
thing appears where there was nothing of the kind before. 

We thus see that there is a close analogy, in all important particulars, between the 
state of unrest which exists in a supersaturated saline solution before crystallisation 
commences and that indicated by the flickering striae in a supersaturated gas before 
liquefaction takes place. 

* "Chemical and Physical Notes," by J. Y. Buchanan, F.R.S , The Antarctic Manual, 1901, p. 97. 
TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 26 



202 MR J. Y. BUCHANAN ON THE 

Section XVI. — The Determination of the Specific G-ravity of the Crystals of 
A Soluble Salt by Displacement in its own Mother-Liquor, and the 
Volumetric Relations between the Crystals and the Mother-Liquor 

WHICH ARE established BY THE EXPERIMENT.* 

§ 120. The work on the specific gravity of dilute solutions at 19 "5° C. reported in the 
early part of this memoir was interrupted by the arrival of the great anticyclone or heat- 
wave of the summer of 1904, during which observations at a temperature of 19"5° were 
quite impossible. Indeed, the temperature of the laboratory, whether by night or day, 
hardly ever fell below 23° C. or rose above 25° C. It persisted over Northern Europe 
for nearly six weeks, and produced tropical conditions, which were evidenced alike by 
the high temperature of the air and by its insignificant diurnal variation. 

In these circumstances I decided to make use of the time by putting into practice 
a method of determining the specific gravity of soluble salts which I had long 
intended to try. I took it up at first merely as a tour de force in experimentation 
with which to occupy myself during the hot weather, but it turned out to be a valuable 
method of research, and the duration of the spell of hot weather enabled me to prove 
and to use it. 

The specific gravity of an insoluble substance is determined by the amount of 
distilled water which a known weight of it displaces. In the case of soluble salts it 
has been the custom to replace the water by a hydrocarbon or mineral oil. The 
objections to the use of this liquid are numerous, especially when the salt, the specific 
gravity of which it is desired to determine, is rare or costly. Moreover, to judge by 
the want of agreement among the values of the specific gravity of the same salt found 
by different chemists, there is greater uncertainty about the numerical results than 
there should be. One reason for this may be that the salts are not insoluble, but 
only sparingly soluble in the oil, and that sufficient attention has not been given 
to this point. 

There is one liquid in which every soluble salt is quite insoluble, and that is its 
own mother-liquor at the temperature at which the one parted from the other. By 
immersing the salt in its own mother-liquor at the temperature of what we may call 
its birth, and by making the maintenance of this temperature a conditio sine qud non 
of every manipulation during which the two are brought together again, errors due 
to uncertain solubility are eliminated, and contamination of valuable preparations is 
avoided. It is therefore by the immersion of each salt in its own mother-liquor that 
I determined its displacement ; and this, combined with the weight of the salt and the 
specific gravity of the mother-liquor, gave the specific gravity of the salt. 

* This formed the subject of a paper which was read at tlie meeting of the Chemical Society of London on 6th 
April 1905, but it was not published by the Society. I owe it to the courtesy of Professor E. S. Dana that the 
liospitality of the pages of the American Journal of Hcience was extended to it. It appeared in the January 
number of 1906, vol. xxi. jx 25, under the title: — "Ou a Method of Determining the SiJecific Gravity of Soluble 
Salts by Displacement in their own Mother-Liquor ; and its Application in the case of the Alkaline Halides. By 
J. Y. Buchanan." 



SPECIFIC GRA\^ITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 203 

It is obvious that the method is applicable only to salts which have a mother- 
liquor, such as KCl ; EbBr ; CaClaeHaO ; BaCl22H2U ; it is inapplicable to salts such as 
CaClz ; BaClg ; and the like, which have no legitimate mother-liquor. 

§ 121. It is an essential condition of success that the work be carried on in a room, 
for the time being, especially devoted to the purpose, and occupied by one investigator. 
He must have in it everything that he requires, including his balance. The window of 
the room must face the north, and the precautions generally to be observed are similar 
to those prescribed by Bunsen for the practice of his original gasometric method. 

The salts used in this research were jhe chlorides, bromides, and iodides of 
potassium, rubidium, and caesium. The rubidium and caesium preparations were from 
the works of Schuchardt in Goerlitz, and on examination proved to be of the highest 
degree of purity. The potassium salts were also unexceptionable as regards quality, 
and were supplied by Merck. All of these salts dissolve easily, and most of them 
abundantly, in water. They also crystallise with great readiness. 

§ 122. The first operation is to prepare a hot solution of the salt such that, after stand- 
ing over nighb, or for such length of time as may be deemed sufficient, it shall furnish 
about 60 c.c. of mother-liquor and about 1 5 c.c. of crystals. In the case of the potassium 
salts there was no difficulty, as their solubility at all temperatures is well known. The 
solubility of the rubidium and caesium salts had to be determined, at least approxi- 
mately, in each case, in order to economise the costly material. The following simple 
method furnished the required information easily and expeditiously. A suitable vessel, 
beaker or flask, is weighed empty, and then with 25 grams of distilled water, of the 
temperature of the air. The salt is then gradually added and the mixture stirred with 
the thermometer. In the case of every one of these salts the temperature falls rapidly 
while dissolving, and by as much as from 15° to 20°. The salt is added as rapidly 
as it is taken up by the water. When the fall of temperature slackens, a minimum 
is soon reached, while some salt still remains undissolved at the bottom of the vessel. 
It is then continually stirred ; the temperature rises slowly while the salt gradually 
passes into solution, until, at a certain temperature, the amount of salt remaining 
undissolved is such that a further rise of one degree of temperature will evidently 
cause it to disappear. The vessel is now weighed, and, as result, we have the weight 
of salt dissolved in 25 grams of water at about the last observed temperature. AVith 
a little care it is easy to arrange that this temperature shall be in the neighbourhood 
of that of the air. The vessel with its contents is now heated, and salt added by 
degrees, while the temperature rises and finally reaches the boiling point or whatever 
other temperature may have been determined on. Salt is added until the liquid is 
saturated at this temperature. The vessel is again weighed and the salt dissolved 
at the higher temperature is ascertained. These simple experiments, which are com- 
pleted in very few minutes, furnish all the information that is required for the 
economical employment of the material. In the absence of more detailed information, 
the following results obtained in the above way are worth quoting : — 



204 



MR J. Y. BUCHANAN ON THE 



100 Grams of Water Dissolve 



Grams 


98 


164 


264 


225 


51 


157 


93 


121 


156 


222 


of 


RbBr 


KbI 


Rbl 


CsCl 


Csl 


Csl 


CsBr 


CsBr 


CsBr 


CsBr 


at°C. 


12 


20 


boiling 


25 


12 


107 


7-5 


24-5 


50 


93-5 



With this information there is no difficulty in preparing the sohition which shall, 
after allowing for unavoidable loss in preparation, give the required amounts of mother- 
liquor and of crystals. The water is warmed and the pure salt is added while the 
temperature is raised to that of ebullition, or to any lower temperature that may have 
been selected. When the salt has all passed into solution, the liquid is poured into a 
flat crystallising dish and crystallisation begins immediately. The area of the dish 
should be such that the layer of solution shall not be more than half a centimetre thick. 
The mother-liquor is then everywhere in close touch with the crystals. The dish is 
then put away in a cupboard for the night. 

In the morning, the temperature of the contents of the crystallising dish and that 
of the air are taken very carefully. The mother-liquor is then poured ofi" clear into 
a stoppered bottle, while the crystals are collected, allowed to drain, and dried in the 
ordinary way. The temperature which the mixture had when separated is noted as 
that at which the crystals and the mother-liquor are in equilibrium ; and it is exactly 
at this temperature that they have to be brought together again in order to determine 
the specific gravity of the salt. It is at this temperature also that the specific gravity 
bottle is weighed when filled with distilled water and with mother-liquor respectively. 
In fact the temperature of equilibrium and of separation is the only temperature used. 

§ 123. In Table I. the experimental details are given in full in the case of one salt, 
namely, caesium chloride. For the other salts the results only are given, and they are 
collected in Table II. 

All the weights as given represent the weight in vacuo. 

The specific gravity bottle which was used was one of the common and convenient 
form which has a thermometer for a stopper and a lateral capillary tube for the adjust- 
ment of level. Its nominal capacity was 50 cubic centimetres. On three occasions 
one of 25 c.c. capacity was used for determining the displacement of the mother-liquor. 

The molecular concentration (m) of the mother-liquor is determined by titration with 
tenth-normal silver nitrate solution. This solution was made with the greatest care and 
contained exactly 17 grams of silver nitrate in one litre, at the ordinary temperature 
of the laboratory at the time. The burette used was divided into tenths of a cubic centi- 
metre and had a capacity of 50 c.c. The determination of the halogen was not made 
until the specific gravity had been determined, and, if the concentration was not already 
known within narrow limits, a preliminary titration was made, after which the volume 
of mother-liquor was weighed which would certainly require 40zbl c.c. for titration. 
The capacity of the burette from to 40 c.c. was determined by weight with great 
care. The concentration is stated in gram-molecules salt per 1000 grams of water. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 205 



Table I. — Experimental Details in the case of Caesium Chloride. 



Formula and molecular weight of salt 

Temperature ........ 

Mother-Liquor. 

Determination of Specific Gravity. 

Weight of specific gravity bottle .... 
Weight of specific gravity bottle filled with distilled water 
Weight of water which fills it, w^-w-^ 
Weight of sp. gr. bottle + mother-liquor 
Weight of mother-liquor, w^-w-^ 

Specific gravity of mother-liquor, — . 

Analysis. 

Weight of mother-liquor taken ........ 

Cb. cents, ^jy -A-gNOg solution used ....... 

Weight of salt equivalent to silver used, ^ .... 

Weight of water in W gms. mother-liquor, Wg - tv^ . 

Concentration of mother-liquor expressed in gm. -molecules salt per 

thousand gms. water, 0*1— '^ . . . . - . 

Salt in Crystal. 

Determination of Specific Gravity. 

A. Weight of first portion of salt 

Weight of sp. gr. bottle + salt + mother-liquor 
Weight of mother-liquor, w-^-^ - (wj + Wjq) 



Weight of water displaced by mother-liquor 
Weight of water displaced by salt, w^ - Wjg 



W-, 



•^12 



Specific gravity of salt, -^ . 

B. Weight of second portion of salt . 

Sum of weights of the two portions, w-^q + w^^ 

Weight of sp. gr. bottle -1- salt + mother-liquor 

Weight of mother-liquor, w^^- (wi + w^^) 

w 
Weight of water displaced by the mother-liquor, -^ 

Weight of water displaced by the salt, Wg - w^^ 

Specific gravity of salt, -^^ .... 

C. Weight of water displaced by salt, Wgj - w^^ . 
Specific gravity of salt, — ^ . 

Accepted specific gravity of salt 



ME 
T 






tV- 



13 



Di 

*<'is 



w. 



19 



D„ 



I>3 
D 



CsCl = 168-5 

23-r c. 



38-8900 gms. 
89-2399 „ 
50-3499 „ 
135-0620 „ 
96-1720 „ 

1-9101 



1-0334 gms. 
41-21 c.c. 

0-6944 gms. 
0-3390 „ 

12-1563 



22-1229 gms. 
U6-5514 „ 
85-5385 „ 

44-7828 „ 

5-5671 „ 

3-9739 

26-6220 gms. 
48-7449 „ 
160-4249 „ 
72-7900 „ 

38-1085 „ 
12-2414 „ 

3-9820 

6-6743 gms. 

3-9890 

3-982 



For weighing out the salt and passing it directly into the specific gravity bottle a 
special and convenient form of weighing tube was used. It was made out of a stoppered 
specimen tube with an internal diameter of 2 centimetres and a length of 7 or 8 centi- 
metres. The lower end of this tube was opened and a piece of narrower glass tube 
joined to it before the blowpipe. This tube, which had a length of about 3 centimetres, 
had an external diameter such that it could just pass freely through the neck of the 



206 MR J. Y. BUCHANAN ON THE 

specific gravity bottle. The wide end was closed with a glass stopper, and the narrow 
end with a small india-rubber cork. 

It was the custom to work so as to have about 15 c.c. of dry salt to be added in two 
charges to the specific gravity bottle. These charges were intended to be nearly, 
though not quite, equal. The available supply was distributed between two weighing 
tubes by approximate weight, after which the exact weight of each portion was deter- 
mined in the usual way. The two portions of caesium chloride weighed respectively 
22'1229 and 26"6220 grams, so that in the first determination of specific gravity 
22*1229 grams and in the second 487449 grams were concerned. It is not immaterial 
whether the first portion is charged into the empty specific gravity bottle and the 
mother-liquor poured over the dry powder, or is charged into the bottle which is already 
about half full of mother-liquor. In the former case the elimination of the entangled 
air is difficult and takes time, during which it is not easy to prevent the temperature 
getting out of hand. By the latter process very little air is carried past the surface of 
the liquid and very little stirring with the thermometer, which is required on other 
grounds, suffices to eliminate it. 

§ 124. Owing to the readiness with which these salts crystallise and to the slowness 
with which all salts dissolve in an almost saturated solution, the temperature of the 
mixture of salt and mother-liquor, during the adjustment of level in the specific gravity 
bottle, must on no account be permitted to fall below T by even 001", nor should it be 
allowed, even momentarily, to rise above it by more than O"!". The regulation of 
temperature was effected entirely with a standard thermometer divided into tenths 
of a degree, each tenth occupying a length of rather more than one millimetre on 
the stem. The thermometer which forms part of the specific gravity bottle is used 
chiefly as a stopper of convenient form. So soon as the level of the liquid has been 
adjusted in the bottle, it is weighed. The temperature and pressure of the air are 
kept account of for the reduction of all weights to the vacuum. 

When the first weighing has been completed, about 20 or 25 c.c. of the clear mother- 
liquor are drawn off and the second charge of dry salt is added and mixed, after winch 
the level is adjusted, and the weight determined. In the absence of experience it might 
be thought that it would be difficult to draw oft' so much of the liquid without some of 
the solid salt ; but no matter how much they may be stirred up, these crystallised salts 
settle at once and completely to the bottom when immersed in their saturated solutions, 
and the operation presents no difficulty. It was at first intended to make a series of 
three determinations with each salt, but two were found to be sufficient. During all 
these manipulations the temperature of the air in the laboratory never differed from 
that of crystallisation (T = 23'l°) by more than one or two tenths of a degree, and, 
when the solubility of the salt is great, it is only in such conditions that operations 
of this kind can be carried out successfully. 

Before bringing the crystals together with the mother-liquor in the specific gravity 
bottle, the operator must realise that their common temperature ivhen mixed is to be 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 207 



exactly that of crystallisation or equilibrium (T) ; and he must take such m^easures 
as his experience dictates to ai-rive at this end. Preliminary experiments on a some- 
ivhat extensive scale are absolutely necessa.ry, and the success of an operation depends 
almost entirely on the operator and the trouble that he is prepared to take. 

§ 125. Table II. gives for each salt, MR, the temperature, T, of equilibrium between 
crystals and mother-liquor, and, in condensed form, the experimental data of the 
determination of S, the specific gravity at T of the mother-liquor, that of water at the 
same temperature being unity, of m, the concentration of the mother-liquor in gram- 
molecules salt per 1000 grams of water, and of D^, Dg, D3, the three observed values, as 
well as D, the finally accepted value, of the specific gravity of the salt, all at T, and 
referred to that of water at the same temperature as unity. 

Table II. — Experimental Results regarding each Salt in the Ennead. 



Salt ; Formula 

MR 
Salt : Mol. weight . 
Temperature, T 



Specific Gravity. 

Weight taken, gms. Wj 
Displacement, gms. jWg 

Specific gravity, — § = S 



KCl. 

74-6 
23r 



KBr. 

1191 
23-4° 



KI. 

166-1 
24-3° 



EbCl. 

121-0 
22-9° 



EbBr. 

165-5 
23-0° 



Rbl. 

212-5 
24-3° 



CsCl. 

168-5 
23-1° 



CsBr. 

213-0 

21-4' 



Csl. 

260-0 
22-8* 



Mother-Ltquor. 



59-4068 
50-3524 

1-1798 



Concentration. 
Gm.-mols. p. 1000 gm. H2O, m 4-7619 



Specific Gravity. 

A. Weight of first portion of 

salt, gms. H'jQ . 
Displacement, gms. w.^^ 

Specific gravity J2=Dj 

B. Weight of both portions of 

salt, gms. Wjg . 
Displacement, gms. m-,,^, 

Specific gravity — i^ = D„ 

C. Weight of second portion of 

salt, gms. xv^^r, ■ 
Displacement, gms. w^^ - w^^ 

= ^21 • . . . 

Specific gravity, —^ = D3 



Accepted specific gravity, D 



34 3044 


85-9636 


74-7356 


81-3282 


46-2696 


96-1720 


42-3756 


24-9554 


49-9140 


49-9188 


49-9196 


24-9478 


50-3499 


24-9744 


1-3746 


1-7222 


1-4971 


1-6292 


1-8548 


1-9101 


1-6968 


57250 


8-9344 


7-7670 


6-7229 


8-2307 


12-1563 


5-3057 



78-0087 
50-3658 

1-5488 



3-5454 



Salt in Crystal. 



13-3684 


36-7928 


27-1751 


19-0112 


27-0906 


26-4777 


22-1229 


27-8926 


7-3271 


13-7498 


8-9703 


7-0256 


8-4700 


7-7248 


5-5671 


6-2453 


1-8245 


2-676 


30295 


2-706 


3-198 


3-428 


3-974 


4-466 


27-4258 


52-5142 


52-1768 


43-7750 


51-5438 


50-6025 


48-7449 


57-5390 


14-5322 


19-6005 


17-1465 


15-9627 


16 0568 


14-7658 


12 2414 


12 9466 


1-887 


2-679 


3043 


(2-74) 


3-210 


3-428 


3-982 


4-455 


14-0574 


15-7214 


25-0017 


(24-76) 


24-4532 


24-1248 


26-6220 


29-6464 


7-2051 


5-8501 


8-1762 


8-9371 


7 5868 


7-0410 


6-6743 


6-7013 


1-951 


2 688 


3-058 


(2-77) 


3-223 


3-426 


3-989 


4-424 


1-951 


2-679 


3-043 


2-706 


3-210 


3-428 


3-982 


4-455 



26-3890 
5-8545 

4-5075 



53-3916 
11-8423 

4-5085 



27-0026 

5-9878 
4-509 

4-508 



208 MR J. Y. BUCHANAN ON THE 

The letters and suffixes have the same significance as in Table L 

The numbers in line T show how uniform the temperature was during the period 
over which the experiments were spread. All the experiments were made between the 
12th and 22nd of July 1904, with the exception of those on caesium bromide, which 
were made on 10th August. By that time the anticyclone had begun to break, and 
the value of T for this salt is 2r4°. For all the other salts, T lies between 22"8'' 
and 24-3°. 

During the whole of the period the barometer was very steady, varying between 
758 and 761 millimetres, and the relative humidity of the air in the laboratory varied 
between 40 and 50 per cent. 

Of the three values D^, Dg, Dg for the specific gravity of the salt, Di is obtained 
directly from the first portion of the salt, Dg from the sum of the two portions, and 
Dg is derived from D^ and Dj by subtraction, 

D.2 represents very nearly the mean of D^ and Dg, and is the accepted value for the 
majority of the salts. It is expressed to three places of decimals, of which units in the 
second place are exact. 

It will be noticed that in the case of rubidium chloride the value of D^ is accepted. 
The second determination depends on the approximate weight of the second portion of 
salt when the tube was being filled, the exact weighing on the balance of precision 
having been accidentally omitted. The operation was however completed, and the 
calculation made with the approximate weight was used as a control. The result shows 
that the value of l)i may be safely accepted. In the case of potassium chloride the value 
of Dg (1'951) is accepted, and the reason for this is as follows: The first portion of 
salt was in very coarse powder, and in mixing it with the mother-liquor numerous 
crystalline particles were observed which contained gaseous enclosures, easily per- 
ceptible by the naked eye. As was expected, the observed specific gravity proved 
to be low. The second portion was much more finely powdered and the specific 
gravity resulting from the two was higher (1"887). But this result is affected 
to the full extent by the gaseous enclosures in the first portion. We therefore 
calculate the specific gravity from the second portion alone, which gives 1"951 for 
the specific gravity. 

It is an advantage of the method just described that it furnishes more than the 
mere determination of the specific gravity of the salt. Thus, by ascertaining almost 
simultaneously the specific gravity of the mother-liquor and the displacement in it of 
the crystals, both being at the temperature of equilibrium, data are obtained for the 
determination of the relation between the displacement of the salt in crystal and the 
increment which it produces in the displacement of 1000 grams of water when it is 
dissolved in this mass of water and forms a saturated solution with it at that 
temperature. It has not hitherto been permissible to make exact comparisons of this 
kind, owing to the independence of the observations on the salt and on the solution 
which have been available. 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 209 

S 126. In discussing the results of observation it is convenient to arrange them in a 
more articulate form than that of Table II. 

The group of salts which forms the subject of these experiments is one of the most 
remarkable in nature. The salts are nine in number and include all the possible binary 
combinations of the members of the electro-positive triad K, Rb, Cs with those of the 
electro-negative triad CI, Br, I. The two triads of simple bodies make three triads, 
or one ennead, of binary compounds. The relations of the different members of the 
ennead are shown in Table III,, in which the different features of the salts are exhibited 
in separate sub-tables. In these sub-tables the data referring to salts of the same 
metal (M) are found in the same column under the symbol for the metal (K, Rb, Cs), 
and those relating to salts of the same metalloids (R) are found in the same line opposite 
the symbol for the metalloid (CI, Br, I). The symbol MR is used to represent both 
the formula and the molecular weight of the salt. 

Sub-table (a) of this table contains the formula and sub-table (c) the molecular 
weight of each salt. The latter is the fundamental attribute of a substance, on which 
all its properties depend. The molecular weights of the salts which occur in one 
column differ by the amount of the difference of the atomic weights of the metalloids 
which they contain, that is, by 44 '5 or 47. Similarly, contiguous salts in one line have 
molecular weights which differ by 46 "4 or 47 '5. If we consider the two diagonal triads 
in the ennead, we see that they are characterised by the fact that both the elements 
in each unit are different from those in either of the other units. Further, along the 
diagonal KCl-CsI the molecular weights of the units differ as much as possible from 
each other, while the atomic weights of the components of each unit are as nearly as 
possible identical, being close neighbours in the atomic series. On the other diagonal, 
KI-CsCl, the molecular weights of the units agree with each other as nearly as possible, 
while the atomic weights of the constituents of the units differ from each other as much 
as possible. 

In sub-table (b) we have the values of T, the temperature at which the crystals 
and mother-liquor of each salt were in equilibrium, and that at which the various 
displacements were observed. 

Under the experimental conditions, which have been minutely described above, it 
is impossible to fix in advance the exact temperature of equilibrium of the crystallising 
liquid. This is given by the meteorological conditions, modified by the structural 
features of the laboratory and of the apartment or enclosure where crystallisation 
takes place. 

§ 127. The Crystal. — In compartment (g) we have the values of D, or the specific 
gravity of the salt in crystal at T, referred to that of distilled water of the same 
temperature as unity. The data in this compartment are in most cases for different, 
but always neighbouring, temperatures. The differences of the values of T are, how- 
ever, so small and those of D are so great that we may discuss the specific gravities 
as if they had been made at one common temperature. 

TEANS. KOY. SOC. EDIN., VOL. XLIX., PART 1. (NO. 1). 27 



210 



MR J. Y. BUCHANAN ON THE 



Table III. — Table giving Numerical Relations between the Crystallised Salts 
of the Ennead MR and their Mother- Liquors. 






K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


K. 


Rb. 


Gs. 




(6) The common temperature at 










which the determinations of the 










(«) Formula of each salt. 


specific gravity of the crystals 


(c) Molecular weight of each salt. | 






and the mother-liquor respect- 












ively were made. 










MR. 


T. 




MR. 




CI. 


KCl. 


RbCl. 


CsCl. 


23°4 


22-9 


23-1 


74-6 


121-0 


168-5 


Br. 


KBr. 


EbBr. 


CsBr. 


23-4 


23-0 


21-4 


119-1 


165-5 


213-0 


I. 


KI. 


Rbl. 


Csl. 


24-3 24-3 


22-8 


166-1 


212-5 


260-0 




{d) Weight of salt per 1000 grams of 


(e) Concentration of the mother- 


(/) Weiiiht 


of the mass of the 




water in each solution. 


liquor expressed in gram-mole- 


mother- 


liquor which contains 






cules salt dissolved in 1000 


1000 ffiams of water. 








grams of water. 










w. 


MR='^- 




1000 -fw = W 




CI. 


355-24 


939-81 


2048-34 


4-7619 


7-7670 


12-1563 


1355-24 


1939-81 


3048-34 


Br. 


681-85 


1112-64 


1130-11 


5-7250 


6-7229 


5-3057 


1681-85 


2112-64 


2130-11 


I. 


1484-00 


1749-02 


921-80 


8-9344 


8-2307 


3-5454 


2484-00 


2749-02 


1921-80 




[g) Specific gravity of the crystal at 


{h) Specific gravity of the mother- 


{i) Displacement of W 


grams of 




T, referred to that of distilled 


liquor at T, referred to that of 


mother- 


liquor at T, expressed m 1 




water at the same temperature 


distilled water at the same 


grams of water at T. 






as unity. 


temperature as unity. 




W 






D. 


S. 




1 = ^- 




CI. 


1-951 


2-706 


3-982 


1-1798 


1-4971 


1-9101 


1148-70 


1295-71 


1595-90 


Br. 


2679 


3-210 


4-455 


1-3746 


1-6292 


1-6968 


1223-52 


1296-73 


1255-37 


I. 


3-043 


3-428 


4-508 


1-7222 


1-8548 


1-5488 


1442-34 


1482-11 


1240-83 








{I) Mean increment of displacement 




{j) Displacement of one gram-molecule 


[k) Increment of displacement of 1000 


of moti 


ler-liquor per gram-mole- 




of the crystal at T, expressed in 


grams of water caused by the dis- 




)f water at T 






grams of water at T. 


solution in it of m . MR. 










MR 


^-1000=% 




A -1000 V 






D 






m m 




CI. 


38-233 


44-710 


42-310 


148-70 


295-71 


595-90 


31-223 


38-069 


49-021 


Br. 


44-460 


51-553 


47-820 


223-52 


296-73 


255-37 


39038 


44-131 


48-137 1 


I. 


54-580 


61-986 


57-670 


442-34 


482-11 


240-83 


49-506 


58-575 


67-907 






{n) Difference of the molecular dis- 










{m) Displacement of one gram- 
molecule of crystal, expressed in 
gram-molecules of water at T. 


placement of the salt in crystal 
from the mean molecular incre- 
ment of displacement of the water 
in the mother-liquor. 




(o) Ratio. 






MR 


MR V 




MR m 






18D" 


D m- 




D ■ V 




CI. 


2-124 


2-489 


2-350 


7-010 


6-641 


- 6-711 


1-225 


1-175 


0-863 


Br. 


2-470 


2-864 


2-657 


5-422 


7-422 


- 0-317 


1-139 


1-168 


0-993 


I. 


3-032 


1 3-444 


3-204 


5-074 


3-411 


-10-237 


1-103 


1-059 


0-849 i 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 211 

On examining the values of D, we see that they increase with those of ME, 
but the increase is not continuous, it is remittent. It takes place triadwise ; and 
this holds whether we take the triads in column or in line. Comparing salts in the 
same line, we see that replacing Rb by Cs causes a rise of specific gravity which is 
twice as great as that caused by the substitution of Eb for K. Comparing salts in 
the same column, the replacement of CI by Br causes more than double the rise caused 
by the substitution of I for Br. However we regard it, we see that the specific gravity 
of the salts is a periodic function of their molecular iveight, within the ennead. 

In compartment {j) we have the values of :^ or the displacement of one gram- 

molecule (ME) of salt stated in grams of water, and in compartment (m) the same 

constant is stated in gram-molecules of water f — = ). In dealing with the specific 

gravities, we saw that, whether we follow the columns or the lines, they increase 
with increase of molecular weight. In the case of the molecular displacements this 
holds for the columns but not for the lines. In these the salts of rubidium have 
the greatest molecular displacement, the potassium salts have the least, and the 
caesium salts occupy an intermediate position. As we shall see later, this irregularity 
is due to a specific peculiarity of the caesium salts. Meantime it may be noted 

that the values of -^^, which may be called the volumetric equivalent of one gram- 
1 oU 

molecule of any of the salts of the ennead, varies from 2124 HgO to 3 "2 04 H2O, the 

iodides having the highest and the chlorides the lowest equivalents. The average 

difference between the volumetric equivalents of the iodides and bromides is 0'563 

H2O, and that between those of the bromides and chlorides is 0'343 HgO. 

§ 128. TJie Mother-liquor. — The values of T are the same for the mother-liquor as 
for the crystals, and are presented in (&). In (e) we have the values of m or the 
molecular concentration of the mother-liquor. This is expressed in gram-molecules salt 
per 1000 grams of water, its equivalent w in grams is given in sub-table {d), and the total 
weight in grams (W) of the solution is given in sub-table {/). The concentration, 
m, of the mother-liquor represents with great exactness the molecular solubility of 
the salt at T, and we shall consider it for a moment from this point of view. 

The least soluble, molecularly, of the nine salts is caesium iodide, which has the 
highest molecular weight, and potassium chloride, which has the lowest molecular 
weight, comes next to it. Next to caesium iodide, in molecular weight and in solubility, 
we Jiave caesium bromide ; and, similarly, next to potassium chloride, in molecular 
weight and in solubility, we have potassium bromide. In the latter case the solubility 
increases with the molecular weight, while in the former it decreases with it. But, if 
sub-table (c) be referred to, it will be observed that, as regards molecular weight, KCl 
and Csl occupy singular positions in the ennead. On the other hand, KBr (119-1) 
and EbCl (121) have almost identical molecular weights, as have also CsBr (213) 



212 MR J. Y. BUCHANAN ON THE 

and Rbl (2 12 '5), yet the solubilities in each pair respectively are very diflFerent. 
The lowest solubilities are on the diagonal KCl- Csl, and the highest solubilities on the 
diagonal KI-CsCl. RbBr, which occupies the middle place on both these diagonals, 
is also in the middle of the middle column and of the middle line, and is the centre 
of the ennead. Its solubility, besides being nearly the average of the group, has a 
symmetrical position with respect to those of the other salts. On one diagonal the 
solubility of its neighbours is lower, on the other higher, than its own. In its column 
the solubilit)^ of its neighbours is higher, in its line it is lower, than its own. 
Turning from the molecular solubilities in sub-table (e) to the ordinary solubilities 
given in sub-table (d), we see that the positions of Csl and KCl are reversed ; the least 
soluble salt of the ennead is KCl, with 355 '24 grams, and next to it comes Csl, with 
921 '80 grams per 1000 grams of water. Other great differences occur which are 
obvious on inspection and need not be further referred to here, because in the research 
only the molecular weights of the salts are taken into account. 

In compartment (h) we have the values of S, the specific gravity of the mother- 
liquor at T, referred to that of distilled water of the same temperature as unity. These 
numbers cannot, as they stand, be compared with each other because they refer to 
solutions of such different concentrations. They enable us, however, to arrive at the 
increment of the displacement of 1000 grams of water caused by its being saturated 
with the particular salt at T. Thus, taking again caesium chloride as an example, we 
have for the weight of salt dissolved in 1000 grams of water 

tv = m.CsCl = 2048-34 grams. 

Adding 1000 grams to this, we have for the weight of the solution 

W = 1000 + «« = 3048-34 grams. 

The specific gravity (S) being r9101, the displacement of the solution is 

W 
A = — - = 1595 '90 grams of water, 
S 

whence the increment of displacement of the water by its saturation with the salt is 

w = A - 1000 = 595-90 grams, 

and the mean increment of displacement per molecule is 

- = 49'021 grams. 

TO 

m. MR + 1000 
Q = A = displacement of the mass of mother-liquor containing 1000 grams of water. 

A- 1000 = V = increment of displacement due to dissolution of m.MR in 1000 grams of water. 

In compartment (/) we have the value of — for each member of the ennead. 

m 

§ 129. Before commenting on the numbers in the table, it is important to form a 
clear conception of their physical meaning. We shall best arrive at this by returning to 
our detailed example of chloride of csesium. As the quantity of saturated solution 
which contains 1000 grams of water weighs 3048 "34 grams and displaces 1595*90 grams 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 213 

of water, we may imagine it to have been prepared in the following way: — 1595*90 
grams of water are taken, and caesium chloride is dissolved in it so that each portion, 
as it is added, forms a saturated solution with the exact quantity of water which, it 
requires for this purpose, and the remainder of the water remains uncontaminated. 
Parallel with the dissolution of the salt, pure water is removed at such a rate as to 
keep the displacement or bulk of the liquid always the same. When no more salt will 
dissolve, we have a saturated solution which contains 1000 grams of water. The weight 
of caesium chloride which has entered the solution is 2048 "34 grams, and the weight 
of water which has left it is 595 "90 grams, whilst the displacement of the liquid is the 
same at the end of the operation as it was at the beginning. In thus describing the 
preparation of the saturated solution, we have described an operation of substitution. 
It is therefore permissible to regard saturated solutions as products of substitution. 
If we give to the above numbers their molecular interpretation, we see that the mean 
increment of displacement produced by the presence of one molecule of caesium chloride 
in its saturated solution at 23*1° is equal to that of 2723 gram-molecules of free 
water, and therefore, that, in these conditions, CsCl is, in a sense, volumetrically 
equivalent to '2 '7 23 H^O. 

If we study sub-table (/), we see that the average molecular increment of dis- 
placement produced by the salts increases with their molecular weight, whether we 
follow the columns or the lines. The only exception is furnished by caesium bromide, 
the increment produced by which is very slightly lower than that of caesium chloride. 
The greatest increment is that due to caesium iodide, which has the highest molecular 
weight ; and the least increment is that due to potassium chloride, which has the 
lowest molecular weight. The pair, potassium bromide and rubidium chloride, which 
have almost equal molecular weights, cause also almost equal molecular increments 
of displacement. The same is true of the pair, potassium iodide and caesium chloride, 
but rubidium bromide causes a markedly lower increment of displacement. Finally, 
the pair, rubidium iodide and caesium bromide, which have almost identical molecular 
weights, present no resemblance in the increment of displacement which they produce. 

§ 130. Comparison of the Displacement of the Salt in Crystal and the Increment 
of Displacement ivhich it produces in the Water of its Mother -Liquor. — The molecular 

displacement .j= of the salts in crystal is given in sub-table (_;* ) in terms of grams 

of water ; that of — , the salts in mother-liquor, is similarly given in sub-table {I). 

If we compare these two tables, we find the remarkable result that while in the case 
of the potassium and the rubidium salts the numbers for the displacement in crystal 
are greater than those for the increment of displacement in mother-liquor, in the case 
of the caesium salts the reverse is the case. 

In sub-table in) we have the diiference ( ^^^ ) of the molecular displacement of 

V D mj 



214 MR J. Y. BUCHANAN ON THE 

the salt in crystal from its mean molecular increment of displacement of the water in 

/MR m\ 
the mother-liquor. In compartment (o) we have the ratio / ^=- . — j of these quantities. 

Taking the figures in compartment (w), we see that in the case of the salts of 
potassium and rubidium crystallisation is accompanied by considerable expansion, and 
this is what is usually met with. In the case of the caesium salts the reverse is the case, 
and very decidedly so in that of the chloride and of the iodide, but much less so in the 
case of the bromide, which, in this, as in other particulars, maintains its singular position. 

In this connection it should be noted that among the ratios (-i=^ ,— ) given in com- 
partment (o), the two which are nearest to unity are those for Rbl (r059) and for CsBr 
(0"993) respectively ; and their molecular weights are almost identical. Further, the 
salts situated co-diagonally to them, namely RbBr and Csl, have ratios whose differences 
from unity are, numerically, almost equal, namely + 0"168 for RbBr and — 0'151 for Csl. 

Taking a general view of the numbers in (o) which give the ratios of displacement in 
crystal and in mother-liquor, we see great differences. The most striking examples are, 
as in the case of solubility, the extreme members of the ennead, KCl and Csl. The 
former expands by more than 25 per cent., and the latter contracts by 15 per cent, on 
crystallising. 

These figures accentuate the peculiarity of the csesium salts, that crystallisation is 
accompanied by contraction. An interesting conclusion can be drawn from the 
behaviour of the different salts in this respect, namely, that the crystallisation of the 
potassium and rubidium salts of the ennead m^ust he hindered by increased pressure, 
while that of the caesium salts must be helped by the same agency. 

§ 131. Extension of the Research to the Salts of the Ennead MRO^, or the 
Oxyhalides of Potassium, Rubidium, and Caesium. — It appeared to be interesting to 
extend this work so as to include the salts of the ennead of the oxyhalides, having 
the general formula MROg, in which M may be K, Rb, Cs, and ROg may be CIO3, 
BrOg, IO3. 

In contrast with the salts of the ennead MR, which are very soluble, the oxyhalides 
are only sparingly soluble. The determination of the specific gravity of the crystals in 
their mother-liquors is therefore much easier, and was effected quite successfully by my 
assistant, Mr H. F. Fermor. The results so obtained are given in Table IV., which 
is identical in form with Table II., dealing with the salts of the halides, which has 
already been explained. 

§ 132. The results of the discussion of the observations made with the salts of the 
ennead MRO3 are given in Tabic V., which is constructed on the same plan as Table III. 
It consists of a number of sub-tables, (a), {b), (c), etc., and the nature of each is 
specified in its title. The molecular weight of each salt, represented by the general 
formula MRO3, differs from that of the corresponding salt of the general formula MR 
by O3 = 48. Therefore the differences between the molecular weights in the same 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 215 



column and between those in the same line in sub-table (c) are the same as those 
between the molecular weights of the corresponding salts of the ennead MR to be found 
in sub-table (c) of Table III., and what was said in this respect about the linear, 
columnar, and diagonal relations of the molecular weights of the salts of the ennead 
MR applies equally in the case of the ennead MRO3. The concentration, m, of the 
mother-liquor, given in sub-table (e) is derived from its specific gravity. 

5 133. In sub-table {g) we have the values of D, the specific gravity of the salt in 
crystal at T, referred to that of distilled water at the same temperature as unity. 
If we examine the values of D, we see that they rise triadwise and parallel to the 
values of the molecular weight. In order to study their differences the accompanying 
table has been constructed : — 

Table giving the Specific Gravities, D, of the Salts of tJie 
Ennead MRO^, and their Differences. 





K. 


Diff. 


Eb. 


Diff. 


Cs. 


C103 . 


2-319 


0-857 


3-176 


0-406 


3-582 


Diff. . 


0-900 




0-505 




0-527 


BrOg . 


3-219 


0462 


3-681 


0-428 


4-109 


Diff. . 


0-705 




0-655 




0-740 


IO3 . . 


3-924 


0-412 


4-336 


0-513 


4-849 



In this table we have the nine entries of the specific gravity of the crystals, and 
these furnish six entries of independent differences taken column-wise, and an equal 
number taken line-wise. The differences occurring in the lines correspond to pairs 
of salts having the same acid and different bases ; those occurring in the columns 
correspond to pairs of salts having the same base but different acids. In the upper 
left-hand corner we have in the top line 0-857, which is the excess of the specific 
gravity of RbClOg over that of KCIO3, and 0"406, which is the excess of the specific 
gravity of CsClOg over that of RbClOg ; so that 0*857 is the increase of the specific 
gravity of the salt MCIO3 when the substitution of Rb for K as the value of M is 
efiFected. Similarly, the increase of specific gravity caused when the substitution of Cs 
for Rb in MCIO3 is effected, is 0-406. 

Replacing CI by Br as R in KRO3 produces a rise of 0"900 in the specific gravity, 
while the replacement of K by Rb as M in MCIO3 produces a rise of 857. When 
Rb is replaced by Cs as M in MCIO3 and MBrOg the effects are similar, namely, a rise 
of 0-406 and 0*428 respectively. The replacement of Br by I as R in KRO3 and 
CsROg causes a rise of 0-705 and 0-740 respectively, while the replacement of Rb by 
Cs as M in MCIO3 is very close to that produced by the replacement of K by Rb as 
M in MIO3, namely, 0-406 and 0-412 respectively. These examples illustrate the 
similarity of the substitution effect produced by elements having nearly identical 
atomic weights but antagonistic chemical and physical properties. 



216 



MR J. Y. BUCHANAN ON THE 



Table IV. 
Experimental Results regarding each Salt in the Ennead MRO^. 



Salt -. Formula MRO3 
Salt : Mol. weight . 
Temperature, T 



Specific Gravity, 

Weight taken, gms., w^ 

Displacement, gms., w.^ 

w 
Specific gi-avity, — 5 = S 



Coneeniration. 
Gm.-mols. p. 1000 gms. HgO, m 

Specific Oravity. 

A. Weight of first portion of 

salt, gms., Wju 
Displacement, gms., Wj^ 

Specific gravity, -^ = D^ 

B. Weight of both portions of 

salt, gms., ?t'jg 
Displacement, gms., w^^ 

Specific gravity, -^ = Dg 
'"20 

C. Weight of second portion of 

salt, gms., w-^c, 



Displacement, gms., ?<;2 



■ w 



= w„ 



w. 



Specific gravity, -i5 = D3 
Accepted specific gravity, D 



KCIO3 
122-6 
14-8° 



KBrOg 
167-1 
19-2° 



KIO3 
214-1 
18-6° 



RbClOg 
169-0 
16-2° 



RbBrOg 
213-5 

16-0° 



RbI03 
260-5 
15-6° 



CsClOg 

216-5 ■ 

16-0° 



CsBrOj 
261-0 
16-0° 



CsIO, 

308-0 

15-4* 



Mother-Liquor. 



1 

52-2451 


52-7937 


53-9820 


52-1316 


51-2751 


51-6035 


52-4485 


51-4449 


51-3776 


50-4321 


50-3973 


50-4135 


50-4209 


50-4223 


50-4258 


50-4235 


50-4219 

1 


50-4274 


1-0360 


1-0475 


10708 


1-0339 


1-0169 


1-0233 


1-0402 


1-0203 


1-0188 


0-4764 


0-3990 


0-4027 


0-2938 


0-1029 


0-1072 


0-2596 


0-0995 


0-0720 


Salt in Crystal. 




6-5566 


23-5976 


38-2490 


29-0782 


32-6042 


39-0514 


29-0122 


25-4094 


38-9291 


2-8211 


7-2897 


9-7153 


9-1539 


8-8402 


8-9315 


8-0905 


6-1773 


8-0102 


2-3241 


3-2371 


3-9370 


3-1766 


3-6882 


4-3723 


3-5860 


4-1134 


4-8600 


12-3282 


53-0730 


79-6206 


58-6266 


68-1770 


78-6761 


64-9130 


60-7122 


83-3354 


5-3153 


16-4973 


20-2904 


18-4619 


18-5218 


18-1450 


18-1219 


14-7769 


17-1872 


2-3192 


3-2171 


3-9240 


3-1755 


3-6809 


4-3360 


3-5820 


4-1086 


4-8487 


5-7716 


29-4754 


41-3716 


29-5484 


35-5728 


39-6247 


35-9008 


35-3028 


44-4063 


2-4942 


9-2076 


10-5751 


9-3080 


9-6816 


9-2135 


10-0314 


8-5996 


9-1770 


2-3140 


3-2012 


3-9122 


3-1745 


3-6743 


4-3008 


3-5789 


4-1052 


4-8389 

1 


2-319 


3-219 


3924 


3-176 


3-681 


4-336 


3-582 


4-109 


4-849 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 217 



CIO3 

BrOj 

10,. 



CIO3. 
BrOg. 
10,. 



Table V. 

J'ahle giving Numerical Relations between the Crystallised Salts of the 
Ennead MRO3 and their Mother -Liquors. 



K. 



Rb. 



Cs. 



(a) Formula of each salt. 



MRO,, 



KCIO3, 
KBrOg 
KIO,. 



RbClOg, 
RbBrOj 
RblO,. 



CSCIO3 
CsBrOj 
CsIO,. 



Rb. 



Cs. 



(6) The common temperature at 
which the determinations of the 
specific gravity of the crystals 
and the mother-liquor respect- 
ively were made. 
T. 



(d) Weight of salt per 1000 grams of 
water in each solution. 



58-41 
66-67 
86-22 



49-65 
21-97 
27-92 



56-20 
25-97 
22-18 



14-8 
19-2 
18-6 



16-2 
16-0 
15-6 



16-0 
16-0 

15-4 



(«) Concentration of mother-liquor 
expressed in gram-molecules salt 
dissolved in 1000 grams of water. 

w _ 

MRa"'"" 



0-4764 
0-3990 
0-4027 



0-2938 
0-1029 
0-1072 



0-2596 
0-0995 
0-0720 



K. 



Rb. 



Cs. 



(c) Molecular weight of each salt. 



MRO.> 



122-6 
167-1 
214-1 



169-0 
213-5 
260-5 



216-5 
261-0 
308-0 



{/') Weight of the mass of the mother- 
liquor which contains 1000 
grams of water. 

1000-t-it' = W. 



1058-41 
1066-67 
1086-22 



1049-65 
1021-97 
1027-92 



1056-20 
1025-97 
1022-18 



CIO3 

BrOg 

10,. 



CIO3. 
BrO.,. 
10,! 



CIO3 
Br03 
10,. 



((/) Specific gravity of the crystal at 
T referred to that of distilled 
water at the same temperature 
as unity. 

D. 



{h) Specific gravity of the niother- 
li(luor at T, referred to that of 
distilled water at the same tem- 
perature as unity, 
S. 



(i) Displacement of W grams of 
mother-liquor at T, expressed 
in gi'ams of water at T. 



2-319 
3-219 
3924 



3-176 
3-681 
4-336 



3-582 
4-109 
4-849 



1-0360 
1-0476 
1-0708 



1-0339 
1-0169 
10234 



1-0402 
1-0203 
10188 



1021-63 
1018-21 
1014-40 



1015-24 
1004-98 
1004-42 



1015-38 
1005-56 
1003-31 



ij) Displacement of one gram-mole- 
cule of the crystal at T, ex- 
pressed in grams of water at T. 



{k) Increment of displacement of 1 000 
grams of water caused by the dis- 
solution in it of m . MRO3. 

A-1000 = 'y. 



{I) Mean increment of displacement 
of mother-liquor per gram-mole- 
cule of salt dissolved in 1000 
grams of water at T. 

A-1000 _'i) 
m m 



52-867 
51-910 
54-603 



53-212 
58-001 
60-078 



60-441 
63-519 
63-518 



21-63 
18-21 
14-40 



15-24 
4-98 
4-42 



15-38 
5-56 
3-31 



45-399 
45-629 
35-756 



51-858 
48-444 
41-251 



59-264 
55-849 
44-634 



(in) Displacement of one gram-mole- 
cule of crystal, expressed in 
gram-molecules of water at T. 

MRO., 



18D 



(n) Difference of the molecular dis- 
placement of the salt in crystal 
from the mean molecular incre- 
ment of displacement of the water 
in the mother-liquor. 
MR03 _t> 
D m 



(0) Ratio. 

MRO3 m 

D ■ v' 



2-937 
2-884 
3-033 



2-956 
3-222 
3-337 



3-358 
3-530 
3-529 



7-467 

6-281 

18-847 



1-354 

9-557 

18-827 



1-177 

7-670 

18-884 



1-164 
1-137 
1-527 



1-026 
1-197 
1-456 



1-020 
1-137 
1-423 



TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 



28 



218 MR J. Y. BUCHANAN ON THE 

§ 134. In sub-table (j) is given the molecular displacement, MRO3/D, of the crystal 
in grams of water, and in sub-table {m) the same constant MRO3/I8D is given in 
molecules of water. 

In the potassium salts the values of this constant is least for KBrOg, and greatest 
for KIO3. In the rubidium salts there is a progressive increase from the chlorate to 
the bromate and the iodate. In the caesium salts the values for the bromate and iodate 
are identical, and that for the chlorate is only very little lower. 

§ 135. Sub-tables (d), (e), and {/) give the concentration of the mother-liquor for each 
salt, expressed in three different ways. In (e) it is expressed in gram-molecules, m, of 
salt per 1000 grams of water, and for none of them is the value of m as high as 0*5. 
Therefore, although saturated, they cannot be called concentrated or strong solutions. 
As was pointed out in § 132, these values of the concentration of the mother- liquor 
are derived from its specific gravity by extrapolation from the ratios of concentration to 
specific gravity in the most concentrated solutions of the salts, as given in § 26, Tables 
16 to 24. This course was adopted owing to the difficulty of determining analytically 
the concentration of solutions of the salts of the ennead MRO3 and the uncertainty of 
the results obtained by desiccation. The dependence of the value of the concentration 
on that of the specific gravity of the mother-liquor excludes certain lines of discussion 
which were followed in the case of the solutions of the salts of the ennead MR. 

It will be remarked that the specific gravities of the non-saturated solutions were 
all determined at Id'd" C, and are referred to that of distilled water at the same 
temperature as unity, while those of the mother-liquors are determined at temperatures 
inferior to 19*5° C, but the specific gravity of each solution is referred to that of 
distilled water of the same temperature as unity. This almost completely eliminates 
any error in the determination of the concentration of the solution which might accrue 
from the difference of temperature at which the specific gravities were determined. If 
Table 66 in § 28 be referred to, the value of possible error due to this cause can be 
ascertained for the two temperatures 19 '5° and 23° C. The concentration of the 
KCIO3 solution would be given too low by 2 per cent. ; in the case of the other solutions 
the error would be less than 1 per cent. But the specific gravities of the mother- 
liquors were determined at temperature lower than 19*5°, and the error would be less 
and in the opposite sense. 

5 136. In sub-table (m) we have the vakies of — ^^ . They are all positive; 

Dm' 

therefore in every case crystallisation is accompanied by expansion. This is small in 

the case of RbClOs and CsClOg, considerable in that of KCIO3 and the bromates, and 

very high in that of the iodates. It is remarkable that the crystallisation of each of 

the three iodates is accompanied by identical expansion. 

§ 187. Finally, attention must be called to the efi"ect on the molecular displacement 

in crystal of the salts of the ennead MR by the addition of O3 so as to form the 

corresponding salts of the ennead MRO3. 



SPECIFIC GEAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 219 

In the followiug table we have in the first line the values of MR, in the second and 
third lines the molecular displacements MRO3/D and MR/D respectively, in the fourth 
line their differences, in the fifth line their ratios, and in the sixth line the corresponding 
ratios MRO3/MR of their molecular weights. 



Salt. 



MR = 



MRO3 
D 

ME 

D 

MRO3 MR 

D D 

MRO, /MR 



D 
MRO 

mr" 



/MR 
/ D 



KCl. 


KBr. 


KI. 


RbCl. 


RbBi. 


Rbl. 


52-86 


51-88 


54-60 


53-31 


58-00 


60-07 


38-23 


44-46 


54-58 


44-71 


51-55 


61-99 


14-63 


7-42 


0-02 


8-60 


6-45 


-1-92 


1-38 


1-17 


1-00 


1-19 


1-12 


0-97 


1-64 


1-40 


1-28 


1-39 


1-29 


1-41 



CsCl. 



60-44 63-52 



CsBr. 



Csl. 



42-31 

18-13 

1-43 

1-28 



47-82 

15-70 

1-33 

1-22 



63-52 

57-67 

5-85 

I-IO 

1-18 



§ 138. Concluding Remm'ks. — These will be very short. The paper has already 
expanded to an unexpected length, and yet, owing to the enormous amount of ex- 
perimentally established material, the discussion of it which has been possible is far 
from adequate, but an end must be made somewhere. 

The Table of Contents has been drawn up in a form which constitutes it really a 
recapitulation of the principal features of the paper, with reference to the paragraph and 
page where they are to be found, so that the reader has no difficulty in making himself 
acquainted with the matters dealt with in the paper, or in studying those which more 
particularly interest him. This being so, I will content myself by indicating here the 
points about the research which present the greatest interest or novelty. 

Two methods of determining specific gravities are used. Neither of them is new in 
principle, but there are innovations in the details of both. To take the case of the 
determination of the specific gravity of a soluble salt in its own mother-liquor, the 
principle is not new, because, if the common practice of determining the specific gravity 
of a salt in petroleum is adopted, the liquid in which it is weighed is, or ought to be, 
a saturated solution of the salt from which, as a mother-liquor, crystals of the salt can 
be obtained ; but it is obvious that this is a very different case from determining the 
specific gravity of chloride of caesium in its mother-liquor, which contains in solution 
something like two parts of salt to one part of water. To carry out correctly this opera- 
tion, the experimenter must be a trained and very experienced chemist ; but it is not 
necessary to be an experienced chemist to perceive the experimental difficulties of the 
operation ; it is therefore unlikely to be attempted by unsuitable hands. 

The principal method used, namely, that in which the very ancient instrument, the 
hydrometer, is used, also requires to be practised by a trained and experienced chemist 
if it is proposed to obtain results of the exactness recorded in this memoir. But to most 



220 MR J. Y. BUCHANAN ON THE 

people the hydrometer is associated with a rough-and-ready method of ascertaining the 
specific gravity of liquids in public works, and in other similar places, where its use is 
commonly entrusted to a workman ; and the idea of readiness, if not of roughness, is, it 
may be said, habitually associated with the instrument and its use. An important part 
of this paper is devoted to showing how the hydrometer has to be used if the best 
results of which the instrument is capable are to be obtained from it. It will be seen 
that experimental skill and perseverance, and constant attention to many minute pre- 
cautions, are necessary. If this care is taken, the results will be good ; if it is not taken, 
they will be bad. 

When the hydrometric method is practised in the manner here specified, it is possible 
to obtain the specific gravity of liquids with greater accuracy than by any other means. 
Hence, in the case of saline solutions it is possible with it to carry the exact determina- 
tion of the specific gravity of the solutions of a salt to much higher dilutions than is 
possible by other methods. It was to experiment on solutions of such high dilution 
that their specific gravities have hitherto escaped experimental determination, that 
this systematic research was originally undertaken. It will be seen that the results 
obtained fully justify the time and labour expended on them. It has hitherto been 
the general experience that, when two equal quantities of a salt are dissolved seriatim 
in a quantity of water, the diminution of the total volume of the salt and the water 
produced by the dissolution of the first quantity is greater than that produced by the 
further dissolution of the second quantity. It has been proved in this memoir that 
for the solutions of many salts there is a concentration below which this law is reversed. 
It is the first time that this has been unequivocally demonstrated. In the case of 
some salts which, when dissolved so as to furnish solutions of moderate concentration, 
exhibit considerable contraction, they at high dilutions exhibit an expansion, which 
may cause the volume of the solution to exceed the sum of the volumes of the salt 
and the water. 

A similar and very remarkable feature of saturated solutions is shown in the 
case of the salts of the ennead MR. In the saturated solutions of the salts of potassium 
and rubidium the sum of the volumes of the salt and water is greater than that of the 
solution produced, while in the case of the solutions of the caesium salts the reverse is 
the case. From this it follows that increase of pressure must assist the crystallisation 
of the solutions of the caesium salts, and hinder that of the solutions of the potassium 
and rubidium salts. 

The main purpose of this investigation was to determine the specific gravity of 
solutions of moderate concentration and of high dilution. In order to use the same 
hydrometer for these different classes of solutions, its weight was altered by the use of 
accessory weights attached to the top of the stem. It occurred to me during the course 
of the investigation that, V)y carrying this principle further, the use of the hydrometric 
method, in all its delicacy, might be extended to solutions of any degree of concentration 
by increasing the additions made to its weight. It was found that for our hydrometers, 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 221 

closed at the top, solutions having a specific gravity of 1 '2 could be experimented on, 
but the accessory weight required was so great as almost to disturb the equilibrium of 
the instrument. In order to meet this difficulty, the stem of the hydrometer was left 
open, so that the internal weight or ballast could be varied at will. With the open 
hydrometer so constructed, saturated and even supersaturated solutions of very soluble 
salts have been experimented on, and results of the highest interest have been obtained. 

The most noteworthy case is that of calcium chloride in supersaturated solution. 
In it a very remarkable state of unrest was observed before crystallisation took place. 
When the crystallisation of this solution is finished, the sum of the volumes of the 
crystals and the mother-liquor is less than that of the original supersaturated solution. 
The state of unrest which precedes the actual appearance of the first crystal consists in 
a rhythmic series of isothermal expansions and contractions, which cease the moment 
the first crystal appears and heat is liberated. The supersaturated solution exhibits 
veritable symptoms of labour before giving birth to the crystals and becoming itself a 
mother- liquor. The details of this remarkable phenomenon are to be found in Section XV. 

It now only remains for me to discharge the pleasant duty of acknowledging my 
obligations to the gentlemen who have acted as my assistants in the experimental work 
and in the preparation of this memoir. The work has been hard and continuous, having 
extended to nearly ten years, and it is impossible for me adequately to express my 
thanks to these gentlemen for the intelligence, skill, and perseverance with which they 
have all devoted themselves to it. 

The secretarial work connected with it has been very heavy, and it has been managed 
with great ability and success by Mr W. Gr. Royal-Dawson, to whom my best thanks 
are due. The pages of Tables in the memoir will suggest to anyone who is familiar 
with such work the amount of labour which has been expended in their preparation and 
verification. 

The experimental work has for nearly three years been in the hands of Mr S. M. 
BoswoRTH, B.Sc, who has carried it out in a room in the Davy-Faraday Laboratory, 
which was admirably suited to the purpose. My thanks are especially due to Sir James 
Dewar and the Managers of that Institution for their generosity in putting it at my 
disposal. Mr Bos worth's name appears several times in the text in connection with 
some of the more remarkable features chronicled, more particularly in connection with 
the state of unrest occurring in the supersaturated solution of calcium chloride before 
crystallisation. It was owing to his confidence in the exactness of the readings of the 
hydrometer which he observed in this solution, and in the reality of the discrepancies 
which he observed, that the state of unrest was not only noticed but measured. Mr 
BoswoRTH was preceded as my assistant by Mr H. F. Fermor, now of the Metro- 
politan Water Board, to whom a large part of the experimental work recorded in the 
Tables is due. His work was of the highest order, and justified his selection for the 
responsible office which he now holds. Before him, my laboratory assistant was Mr H. 
Royal-Dawson, brother of Mr W. Gr. Royal-Dawson, and he, like all the gentlemen whom 



222 



MR J. Y. BUCHANAN ON THE 



I have been fortunate enough to have as assistants, attained the same high degree of 
exactness in his experimental work, so soon as he perceived that, when he took the 
necessary trouble with the work, it was rewarded by increased accuracy of results. 
This has been my invariable experience. Comparisons of work done on solutions of the 
same concentration of the same salt by Mr Royal-Da wson, and afterwards by Mr 
Bos WORTH, are quoted in the memoir, and they furnish evidence of the excellence of 
the work put out by both these chemists. 

Nearly the whole of the experimental work of this memoir has been done by the 
gentlemen just mentioned. It would, however, be unjust if I did not refer to the great 
and valuable work with the hydrometer done for me at an earlier date by my old and 
valued friend and former assistant, Mr Andrew King of the Heriot-Watt College, 
Edinburgh. The exactness of his work is of the highest order, and his intimate know- 
ledge of, and sympathy with, my work for many years has been of the utmost value to 
me, and I wish to take this occasion to make public acknowledgment of the debt 
of gratitude which I owe to him. 



APPENDIX A. 

Densities of the Solutions at T. 

In the following tables the specific gravities of the solutions have been reduced to 
their value when referred to that of distilled water at 4" C. as unity. The factors 
used for this purpose are : — 

for T= 15-0° 19-5° 23-0° 26-0° 

factor- 0-999173 0-998372 0-997614 0-996879 

For example, 15.815. of ^ NaCl is 1 '020564. Therefore its density 4.8150 = 0-999173 15.815. = 
1-019720. 

CHLORIDES. MCI. 



M = 


Na. 


K. 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


t= 

m. 
1/2 
14 
1/8 
1/16 
1/32 
1/64 
1/128 
1/256 
1/512 


15-0° C. 


19-5° C. 


23-0° C. 


1-019720 
1-009597 
1-004427 
1-001821 
1-000494 
0-999827 
0-999495 


1-022321 
1-011063 
1-005080 
1-002123 
1-000659 
0-999888 
0-999538 


1-021312 
1-010023 
1-004251 
1-001340 
0-999859 
0-999112 
0-998736 
0-9985G5 
0-998454 


1-041446 
1-020204 
1 009377 
1-003903 
1-001139 
0-999770 
0-999078 
0-998721 
0-998535 


1-060842 
1-030059 
1-014340 
1-006395 
1-002400 
1000396 
0-999395 
0-998885 
0-998620 


1-000531 


1 003086 


1-005549 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 223 



BROMIDES. MBr. 



M = 


K. 


Kb. 


Cs. 


K. 


Rb. 


Cs. 


T = 


19-5° C. 


23-0° C. 


m. 

1/2 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/1024 


1-039583 
1-019241 
1-008883 
1-003642 
1-001006 
0-999676 
0-999023 
0-998696 
0-998530 


1-059519 
1029411 
1-013935 
1-006227 
1-002310 
1000326 
0-999354 
0-998828 
0-998605 
0-998450 


1-079175 
1-039316 
1-019040 
1-008764 
1-003545 
1-000999 
0-999640 
0-998978 
0-998679 
0-998517 


1-002907 


1-013316 
1-005490 
1-001597 
0-999577 
0-998598 


1-018237 
1007975 
1-002847 
1-000242 
0-998943 



IODIDES. MI. 



M = 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


Cs. 


T = 


19-5° C. 


23-0° C. 


26-0°C. 


m. 

1/2 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

1/1024 


1-057205 
1-028229 
1-013451 
1-005948 
1-002156 
1-000268 
0-999220 
0-998851 
0-998607 
0-998494 


1-076665 
1-038085 
1-018349 
1-008402 
1003404 
1-000873 
0-999607 
0-998923 
0-998644 
0-998518 


1-096000 
1-048131 
1-023304 
1-010880 
1-004661 
1-001487 
0-999915 
0-999108 
0-998644 
0-998472 


1-056113 

1027260 
1 012568 
1-005140 
1001366 
0-999528 
0-998562 
0-998100 


1-017641 
1-007682 
1-002645 
1000163 
0-998878 
0-998265 


1022635 
1-010221 
1-003940 
1-000769 
0-999206 
0-998526 


1-009463 



NITRATES. M'NOs and M"(N03)2. 



M' or M" 


Na. 


K. 


Sr". 


Ba". 


Li. 


N. 


Ba". 


Pb". 


Rb. 


Cs. 


T = 


15-0° C. 


19-5° C. 


23-0° C. 


m. 
1/2 
1/4 
1/8 
1/16 
1/32 
1/64 
i/128 
1/256 
1/512 
1/1024 


1-002623 
1-000887 
1-000036 
0-999604 


1-030021 
1-014860 
1-007140 
1-003136 
1-001145 
1-000157 
0-999663 


1 004513 
1-001844 
1-000523 
0-999838 
0-999502 


1-005886 
1-002547 
1-000865 
1-000008 
0-999595 
0-999391 


1-018047 
1-008389 
1-003395 
1-000916 
0-999660 
0-999025 
0-998707 


1-026137 
1-012472 
1-005479 
1-001954 
1-000171 
0-999271 


1-011652 
1-005058 
1-001734 
1-000079 
0-999227 
0-998804 
0-998577 


1-016131 
1-007304 
1-002869 
1-000618 
0-999498 
0-998948 
0-998672 


1-023143 

1-010648 
1-004182 
1-000960 
0-999341 
0-998567 
0-998017 


1-015514 
1-006672 
1-002139 
0-999897 
0-998797 
0-998193 



224 



MR J. Y. BUCHANAN ON THE 



R0«= 



Tables giving a Summary of the Densities of the Solutions of different 

Salts at different Temperatures. 

TRIADS OF NITRATES, CHLORATES, BROMATES, AND lODATES. MRO.. 



M= 



T = 



1/2 

1/4 

1/8 

1/16 

1/32 

1/64 

1/128 

1/256 

1/512 

ijie 



N0„ 



K. 



Rb. 



Cs. 



19-5*0. 



028886 
013880 
006232 
002334 
000382 
999374 
998880 



-048924 
•024028 
-011314 
-004958 
-001721 
•000119 
•999290 
-998829 



043933 
016304 
007397 
002950 
000615 
999516 
99)5975 



C10„ 



BrO,. 



K. 



Rb. 



Cs. 



Rb. 



Cs. 



19-5° C. and ^5-0° G. 



19-5° C. and ^5-0° C. 



■017422 
•007994 
-003227 
•000758 
•999623 
•999004 
•998691 




0-998554 



1-002362 



•027478 

•013027 

005716 

•002057 

■000232|l 

•9992890 

■998830 

•9985900 



•037351 1 
■0180261 
-0081811 
-0033171 
-0007771 
•909586 0' 
•9989230' 
•998581 lO' 



030146 
•013570 
•006022 
002212 
000290 
999328 
998847 
998609 



1-004927\1 •007Jf76\l -005 16 J^ 



1-008609 
1-003487 
1-000934 
0^999630 
0^999013 
0-998691 

1-007752 



1 
1 

1-011107 1 
1-0047391 
l-001578|l 
0-999986 



0-999245 
0-998746 

1-010343 



10,, 



Rb. 



Cs. 



19-5° C. and^5^0°C. 



•042602 
-020662 
-009523 
-003952 
-001128 
-999773 
-999080 
-998732 



1-012027 
1-005215 
1-001771 
1-000059 
0-999198 



014644 
006501 
002388 
000317 
999300 
0-9988070-998820 



1 -0087 Sm -011206 1 013801 



STKONG SOLUTIONS {Pyknometer). 

Tables giving a Summary of the Densities of the Solutions of different 
Salts at different Temperatures. 



Ror 
R03 = 


CI. 


Br. 


I. 


NO3. 


M = 
T = 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


K. 


Rb. 


Cs. 


Rb. 


Li. 


Na. 


19-5° C. 


19-5° C, 


21 -4° C. 


19-5° 0. 


23-rc. 


19-5° C. 


m. 
1/2 

1 

2 

3 

4 

5 

6 

7 

8 

9 
10 


1-0432 
1-0835 
1-1204 
1-1543 


1-0409 
1-0814 
1-1573 
12263 
1-2915 
1-3519 
1-4061 
1-4562 


1-0599 
1-1060 
1-2250 
1-3223 
1-4090 
1-4931 
1-5644 
1-6447 
1-6997 
1-7561 


1-0790 
1-1526 
1'2200 
1-2832 
1-3403 


1-0596 
1-1175 
1-2261 
1-3250 
1-4155 
1-4985 
1-5746 


1-1467 
1-3015 
1-4297 
1-5516 
1-6590 


1-1128 
1-2157 
1-3097 
1-3959 
1-4766 
1-5458 
1-6115 
1-6722 


1-0753 
11469 
1-2814 
1-4003 
1-5077 
1-6055 
1 6944 
1-7676 


1-1718 
1-3431 
1-4887 


1-0488 
1-0964 
1-1842 
1-2654 


1-0372 
1-0730 
1-1063 
1-1373 
1-1665 
1-1940 
1-2194 
1-2437 
1-2663 
1-2885 


1-0525 
1-1012 
1-1459 
1-1871 
1-2247 
1-2592 
1-2918 
1-3231 
1-3517 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 225 



APPENDIX B. 

Table giving the Number of Series as well as the Number of Single Observations 
made with the various Hydrometers, from which the results recorded in this 
Memoir were obtained. 



Hydrometer. 




Number 




Number of 
Series obtained. 


of Single 
Observations 




Type. 


Designation. 




made. 


Closed 


No. 3 


391 


3,546 


)> 


„ 17 


1287 


11,583 


J) 


n 21 


511 


4,599 


Open 


A 


207 


2,183 


1? 


B 


164 


1,616 


!> 


1910, No. 3 


22 


198 


Total . 


2585 


23,725 



TRANS. ROY. SOC. EDIN., VOL. XLIX., PART I. (NO. 1). 



29 



226 



MR J. Y. BUCHANAN ON THE 



INDEX. 



Accessory weights, Increase of weight of hydro- 
meter by means of ... . 21 
Multiple observations by means of . . 27 

Preparation of 29 

Andrews, Reference to experiments made on the 
critical temperature and pressure of 
gases by ..... . 201 

Archimedes, Principles of 17 

Beryllium chloride, Comparison of specific 
gravity and displacement of solutions 
of 179-184 

Expansion on dilution of solutions, of low 

concentration, of . . . . . 182 

Specific gravity and displacement of solu- 
tions of 179 

Calcium chloride, Comparison of specific gravity 

and displacement of solutions of . . 179 

Concentration of supersaturated solutions of 185 

Constitution of crystals of . . . . 189 

Eate of cooling of supersaturated solution of 187 

Remarkable state of unrest in a super- 
saturated solution of . . . . 185-196 
Tables of specific gravities of solutions of . 171, 

179, 190-192 
Challenger, Hydrometer used on board the. . 20 

Laboratory on board the .... 24 

Concentration, Change of displacement pro- 
duced in a solution by change of . . 116-131 
Crystallisation, Analogy between formation of 

ice and 199 

Heat liberated during .... 188 

Remarkable state of unrest in a solution of 

calcium chloride preceding . . . 185-196 

Displacement, Difference of . . . 61-72, 102 
Discussion on the values of increments of . 132-144 
Hypotheses used in the discussion of the 

increments of .... . 108 

Increments of .... 102,10.5,135,139 

Mean increment of 105,139 

Statistics of solution 61-72 

Summary of iiiciement.s of . . . . 84-88 
Sunmiary of mean increments of . . 89-93 

Values for difference between increment of 

di-splacement and difference of . 123-129,183 

Ennead, Tables giving relations of the molecular 

displacements of salts of the double . 219 

Table-s giving data relating to salts and 

mother-liquors of the double . . 210, 217 
Exponents, Discussion of relation between . 113-1 16, 184 



Hydrometer, Accessory weights used at the top 

of the stem of .... . 21 
Determination of true weight of . . . 32 
Determination, in distilled water, of dis- 
placement of 33-44 

Determination of specific gravity of solids 
lighter and heavier than water by 

means of 17 

Example of degree of accuracy attainable 

by the use of 172 

Principle and construction of closed . . 26 

Principle and construction of open . . 155 

Study of mineral waters with . . . 19 

Use of accessory weights with ... 20 

Use on board the Challenger of the . . 20 

Hypotheses, Agreement of values of increments 

of displacement for a particular salt 

with one of two 108-116 

Ice, Analogy between the deposition of crystals 

from a solution, and the formation of . 197 
Isomer, Discussion of displacement of solutions 

of the artificial 177 

Isomeric salts, Ditt'erences of displacement . 175 

Solubility of 174 

Specific gravity of solutions of . . . 173 

Use of the term 172 

Laboratory, Challenger ..... 24 

Control of room temperature in . . . 53 

Privacy of 24, 171 

Magnesium chloride, Behaviour of super- 
saturated solution of . . . 178, 196 

Comparison of specific gravity and displace- 
ment of solutions of ... . 179-184 

Specific gravity and displacement of sohi- 

tions of 1 79 

Thermal effects of the dissolution of crystals 

of 179 

Meniscus, Infiuence, on the determination of 

specific gravity, of ... . 46 

Mineral waters, Use of hydrometer in the study 

of 19 

Mother-liquors, Concentration of certain . 187, 207, 217 

Determination of the specific gravity of 

salts by displacement in their own . 202-219 

Discussion relative to 209, 218 

Numerical relation.s between crystallised 

salts and their 209, 218 

Specific gravity of ... . 187, 207, 217 

Tables of dispjlacements of . . . . 207, 216 

Observations, Scheme for logging . . .38, 167 



SPECIFIC GRAVITY AND DISPLACEMENT OF SOME SALINE SOLUTIONS. 227 



Salts, Determination, by displacement in their 

own mother - liquor, of the specific 

gravity of soluble 

Solubility of certain .... 

Use of open hydrometer in determining the 
specific gravity of " isomeric " 
Sodium chloride, Specific gravity and displace 

ment of solutions of . . 
Solids, Hydrometer for the determination oi 

specific gravity of any . 
Solids lighter than water, Hydrometer for 

determination of specific gravity of 
Solution, Behaviour of a supersaturated 
Control of temperature of . 
Determination of the specific gravity of i. 
saline ...... 

Solutions, Displacement of ... 

Increment of displacement of 
Methods of preparation of . 
Series of observations made with hydro 

meter in 

Statistics of displacements of 



203- 


-219 


174- 


-204 




173 


148- 


-154 




17 




17 


185- 


-196 


55-60 





49 




107 


84 


-89 


120, 


149 




44 


61 


-72 



Specific gravity, Determination of . . . 

Discussion of exactness of results of . 

Di.scussion of results of ... . 

Exactness of determination of . 

Influence of meniscus on the determination of 

Summary, in the case of solutions of different 
salts at different temperatures, of 
State of unrest, Large and rapid variations of 
specific gravity in a supersaturated 
saline solution indicating a . 

Phenomena accompanying a 

Supersaturated solution exhibiting a . 

Value of hydrometer to indicate a 

Tables, Displacements of hydrometer in distilled 

water, as given in ... 

Temperature, Example of control of room . 

Importance, in hj^drometric work, of con 

stancy of 

Maximum departure of solution . 
Mean range of solution 
Statistics of range of variation of 



PAGES 

46 

103 

104 

73-79 

46 

81-83 





194 


185- 


-201 


185 


-201 




196 


I 

36-44 


. 55- 


-193 




23 




51 




51 




51 



PRESENTED 

23 JUL 1913 




The Transactions of the Royal Society of Edinburgh will in future be Sold at the! 

following reduced Prices : — 









































Price 


Price 






Price 




Price 






Price 




Price 


Vol. 


to the 


to 




Vol. 




to the 




to 




Vol. 




to the 




to 




Public. 


Fellows. 




Public. 


Fellows. 




Public. 


Fellowa. 1 




£ s. 


d. 


£ s. 


d. 




£ 


s. 


d. 


£ 


s. 


d. 




£ 


s. 


d. 


£ 


<. (L 


VI. 


11 


6 


9 


6 


XXXIII. Pt. 1 


1 


1 








16 





2 . / XLII. 


2 


2 





1 


11 


VII. 


18 





15 





„ Pt. 2 


2 


2 





1 


11 





|§JXLIII. 
H) XLIV.Pt.l 


2 


2 





1 


11 


VIII. 


17 





14 





„ Pt. 3 





12 








9 


6 




18 


6 


1 


9 


IX. 


1 





17 





XXXIV. 


2 


2 





1 


11 





«^l „ Pt.2 




1 








15 9 


X. 


19 





16 





XXXV.*Pt. 1 


2 


2 





1 


11 





XLV. Pt. 1 




9 





1 


2 


XI. 


14 


6 


12 





Pt. 2 


1 


11 





1 


3 


6 


„ Pt. 2 




7 





1 





XII. 


14 


6 


12 





„ Pt. 3 


2 


2 





1 


11 





„ Pt. 3 




13 


9 


1 


5 3 


XIII. 


18 





15 





„ Pt. 4 


1 


1 








16 





„ Pt. 4 





4 


6 





3 6 


XIV. 


1 5 





1 1 





XXXVI. Pt. 1 


1 


1 








16 





XLVI. Pt. 1 




1 


10 





16 6 


XV 


1 11 





1 6 





Pt. 2 


1 


16 


6 


1 


7 


6 


„ Pt. 2 




5 


8 





19 4 


XX. Pt. 1 


18 





14 





Pt. 3 


1 











16 





„ Pt. 3 




7 


3 


1 


11 


XXII. Pt. 2 


10 





7 


6 


XXXVII. Pt. 1 


1 


14 


6 


1 


5 


6 


XL VII. Pt. 1 





19 


9 





15 


„ Pt. 3 


1 ^ 


9 


1 1 





„ Pt. 2 


1 


1 








16 





„ Pt. 2 




3 








17 4 


XXVII. Pt. 1 


16 





12 





„ Pt. 3 





16 








12 





„ Pt. 3 







10 





15 8 


„ pt:2 


6 





4 


6 


„ Pt. 4 





7 


6 





5 


8 


., Pt. 4 




7 


7 


1 


9 


„ Pt. 4 


1 





16 





XXXVIII. Pt. 1 


2 








1 


10 





XLVIII. Pt. 1 




2 


9 





17 2 


XXVIILPt.l 


1 5 





1 1 





Pt. 2 


1 


5 








19 





» Pt. 2 




9 


6 


1 


2 5 


„ Pt.2 


1 5 





1 1 





Pt. 3 


1 


10 





1 


3 

















„ Pt. 3 


18 





13 


6 


„ Pt. 4 





7 


6 





5 


8 














-XXIX. Pt. 1 


1 12 





1 6 





XXXIX. Pt. 1 


1 


10 





1 


3 





XLIX. Pt. 1 





7 


6 





7 6 


„ Pt. 2 


16 





12 





Pt. 2 





19 








14 


6 














XXX. Pt. 1 


1 12 





1 6 





l^t. 3 


2 


3 





1 


11 

















„ Pt. 2 


16 





12 





Pt.4 





9 








7 

















„ Pt. 3 


5 





4 





XL. Pt. 1 


1 


5 








19 





1 












„ Pt. 4 


7 


6 


5 


8 


„ Pt.2 


1 


12 


6 


1 


5 


6 












> 


XXXI. 


4 4 





3 3 





„ Pt.3 


1 


6 








19 


6 














XXXII. Pt. 1 


1 





16 





„ Pt.4 


1 











16 

















„ Pt. 2 


18 





13 


6 


XLI. Pt. 1 


1 


1 








15 


9 














„ Pt. 3 


2 10 





1 17 


6 


„ Pt.2 


1 


9 


6 


1 


2 

















„ Pt.4 


5 





4 





„ Pt.3 


2 


5 





1 


13 


6 















* Vol. XXXV., and those which follow, may be had in Numbers, each Number containing a complete Paper. 



January 1912. — Volumes or parts of volumes nc.*^^ mentioned in the above list are not for the present on sale 
to the public. Fellows or others who may specially desire to obtain them must apply direct to the Society. As 
the Society reprints from time to time parts of its publications which have become scarce, the absolute correctness 
of this list cannot be guaranteed beyond this date. 



TRANSACTIONS 



OF THE 



ROYAL SOCIETY OF EDINBURGH. 

VOLUME XLIX. PAET IL— SESSION 1912-13. 



CONTENTS. 



PAGE 



Antarctic Fishes of the Scottish National Antarctic Expedition. By C. Tate Regan, INI. A., 
-sistant in the British Museum (Natural History). Communicated by Dr W. S. Bruce. 
A'ith Eleven Plates and Six Text-Figs.), . . . . . . .229 

{Issued May 23, 1913.) 

J. mograph on the general Morphology of the Myxinoid Fishes, based on a study of Myxine. 
Irt V. : The Anatomy of the Gut and its Appendages. By F. J. Cole, D.Sc. Oxon., Pro- 
^.sor of Zoology, University College, Reading. Communicated lnj Professor W. A. Hbrdman, 
]R.S. (With Four Plates), . . . . . . . .293 

{Issued June 9, 1913.) 

■ iie Skulls of Antarctic Seals: Scottish National Antarctic Expedition. By "William S. 
BjCE, LL.D., Director of the Scottish Oceanographical Laboratory. (With Five Plates), . 345 

{Issued June 27, 1913.) 

Pohhxta of the Families Serpulidas and Subellidx, collected by the Scottish National Antarctic 
j-pedition. By Helen L. M. Pixell, B.Sc, F.Z.S., Demonstrator of Zoology and Reid 
lUow, Bedford College, University of London. Communicated by Dr J. H. Ashworth. 
('itb One Plate), .......... 347 

{Issued June 24, 1913.) 

Cariocian Cystideafrom Girvan. By F. A. Bather, M.A., D.Sc, F.R.S. Communicated by 
EHoBNK, F.R.S. (With Six Plates.) [Published by permission of the Trustees of the 
Etish Museum], .......... 359 

{Issued July 24, 1913.) 






EDINBURGH: 
PUBLISHED BY ROBKET GRANT k SON, 107 PRINCES STREET, 
ILLIAMS & NORGATE, 14 HENRIETTA STREET. COVENT GARDEN, LONDON 

MDCCCCXIII. 
Price Thirty-two Shillings and Sixpence. 



( 229 ) 



n. — The Antarctic Fishes of the Scottish National Antarctic Expedition. By 
C. Tate Regan, M.A., Assistant in the British Museum (Natural History). Com- 
municated by Dr W. S. Bruce. (With Eleven Plates and Six Text-figs.) 

(MS, received June 18, 1912. Read December 16, 1912. Issued separately May 23, 1913.) 

Our knowledge of the Antarctic fish-fauna has greatly increased during the last 
ten years. The Belgian expedition to Graham Land (1897-1899) was followed by 
that of the Southern Cross to Victoria Land (1898-1900), fitted out by Sir George 
Newnes. Next were the British expedition of the Discovery to Victoria Land and 
Edward Land (1901-1904), the German voyage of the Gauss to Kerguelen and Wilhelm 
Land (1901-1903), and Nordenskjold's Swedish expedition to South Georgia, the South 
Shetlands, and Graham Land. Then came the voyage of the Scotia to the South 
Orkneys and Coats Land (1902-1904), and Charcot's expeditions to the Palmer 
Archipelago and Graham Land in the Fran<^ais (1904-1905) and the Pourquoi Fas? 
(1908-1910), and finally Shackleton's expedition (1908-1909). 

The fishes collected during these expeditions have been described in a series of 
reports, which may be enumerated in chronological order : — 

1902. BouLENGER, Pisces in '■'■Southern Cross" Collections, pp. 174-189, pis. xi.-xviii. 

1904. DoLLo, Res. Voy. " Belgica" : Poissons, 240 pp., 12 pis. 

1905. LoNNBERG, "The Fishes of the Swedish South Polar Expedition," Wisseiiscli. Ergehn. 

Schwedisch. Siidpolar-Exped., v. 6, 69 pp., 5 pis. 

1906. Vaillant, Ex'ped. Antard. Frangaise : Poisso7is, 51 pp. 

1907. BouLENGER, Natwiial Antarctic Expedition, Nat. Hist.: II., Fishes, 5 pp., 2 pis. 

1911. Waite, ^'^ Ania.xcticYis\i&B," \n British Antarctic Expedition, 1907-9: Biology, -p-p. 11-16, 

pi. ii. 

1912. Pappenheim, "Die Fische der Antarktis und Subantarktis," in Deutsche Siidpolar-Exped., 

1901-1903 : XIIL, Zool, v. pp. 163-182, pis. ix.-x. 

Dr DoLLO presented several preliminary notes in the Proceedings of the Royal Society 
of Edinburgh* on the fishes of the Scottish National Antarctic Expedition. 

The fishes of the second Charcot expedition have been worked out by Professor 
Roule, who has published two preliminary notes {C.R. Acad. Sci. Paris, cliii., 1911, pp. 
80-81, and Bull. Mus. Paris, 19 U, pp. 2 7 6-2 81), but the final report has not yet appeared. 

The important collection of fishes here reported on was made at the Falkland Islands, 
the South Orkneys, Coats Land, and Gough Island, and in the Weddell Sea and South 
Atlantic Ocean between these localities. As will be seen from the systematic list that 
follows, it includes examples of forty-eight species, ten of which are now described as 
new to science, whilst three others, known before but wrongly identified, are diagnosed 
and given new specific names ; in addition, four species have already been described by 

* Proc. Roy. Sac. Edin, xxvi., 1906, p. 172 ; xxviii., 1908, p. 58 ; xxix., 1909, p. 316. 
TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 2). 30 



230 



MR C. TATE REGAN ON THE 



Dr. DoLLO, in whose liands the greater part of the collection has been from 1905 until 
March 1912. 

The identification of the Notothenioids and Zoarcids has proved a difficult matter in 
the present state of our knowledge of these groups, and I have supplemented my report 
by a monograph of the former and a revision of the southern genera of the latter ; 
further, I have added some notes on the G-alaxiidse and Haplochitonidse, as their distri- 
bution has given rise to some discussion. 

My work on the Notothenioids and Zoarcids is mainly based on the specimens in the 
British Museum, including the Erebus and Terror, Challenger, Southern Cross and 
Discovery collections, but I have been greatly helped by the loan of specimens from 
the Museums at Paris, Berlin, and Stockholm. Thus I have been able to examine all 
the species of Notothenia recorded by Vaillant from Graham Land, two of the three 
species of Zoarcids recently described by Pappenhkim from Wilhelm Land, and 
co-types of some of the Notothenioids described by Lonnberg. For their kindness 
in sending me these fishes, and in giving me information about others that could not 
be sent, I heartily thank Dr Pellegkin, Dr Pappenheim, and Dr Lonnberg. 

It need hardly be said that the fishes lend no support to the theory of bipolarity. 
Most of the littoral fishes belong to the Nototheniidse and related families, which are 
characteristic of and peculiar to the Antarctic seas and the region immediately to the 
north of them ; there are also several species of Zoarcidse, generically distinct from 
the northern members of the family. Some of the pelagic and abyssal fishes are 
Notothenioids peculiar to the Antarctic region ; others also, such as Notolepis, Cynoma- 
crurus, and Eugnaihosaurus, may not be found elsewhere ; but the rest belong to 
widely distributed genera {Synaphohranchus, Bathylagus, Myctophum, etc.) or even 
species (e.g. Cyclothone microdon). 

In the whole paper the following seven new genera and twenty-one new species 
are described : — 

New Genera. 



Eugnaihosaurus, p. 234. 
0])Iithalmoli/cus, p. 243. 
AustrolyciclitJtys, p. 244. 
Austroltjcus, p. 245. 

Bathylagus glacialis, p. 231. 
Eugnatliosaurus vorax, p. 234. 
Synaphohranchus australis, p. 235 
Chalinura ferrieri, p. 236. 
,, icMtsoni, p. 236. 
Cxstoperca coafsn, ]>. 237. 
Neophrynichthys marmoratus, p. 241. 
Lijcenclielys antarcticus, p. 242. 
Audrolycus depressiceps, p. 245. 
Crossolycus diilensis, p. 247. 
CoUoperca macropJithahita, p. 253. 



New Species. 



Crossolycus, p. 247. 
Pagetopsis, p. 286. 
Chxnoceplialus, p. 287. 



Bovickthys angustifrons, p. 255. 

,, chilensts, p. 256. 

,, decipiens, p. 257. 
Trematomus loennberyii, p. 263. 
Notothenia trigramma, p. 266. 

„ ramsayi, ji. 267. 

,, wiltoni, p. 268. 

., vaillanti, p. 272. 

Chxnichtliys ruyosus, p. 287. 
Cryodraco pappenheimi, p. 289. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 231 

I. Antarctic and Subantarctic Fishes collected by the "Scotia."* 

SELACHII. 
Raid.e. 

1. Raia magellanica, Steind. (PI. I.) 
Znol. Jahrb. SuppL, vi., 1905, p. 212. 
One specimen from Station 346, Burdwood Bank, depth 56 fathoms; taken on 
1st December 1905. f.at. 54° 25' S., long. 57° 32' W. ; temperature 41-8° F. 

This is a female of exactly the same size as Steindachner's type, and apparently in 
every way similar, except that there is only a single scapulary spine, instead of a series 
of three on each side. 

This species is related to R. murrayi, Giinth., from Kerguelen, but has a blunter 
snout, a shorter tail, and somewhat different spination. 

ISOSPONDYLI. 

Clupeid^. 

2. Clupea /uegensis, Jenyns. 

Zool. " BeagI''," Fish., p. 133 (1842); Smitt, Biliang. Svensk. Vet.-Akad., xxiv., 1898, iv.. No. 5, 
p. 59, pi. V. fig. 41 

Depth of body 4 to 5 in the length. Lower jaw very prominent ; minute teeth in 
a single series on the palatines and in an elongate patch on the tongue. Dorsal 17-18 ; 
origin equidistant from anterior edge of eye and base of caudal fin. Anal 17-20. 
Origin of pelvics vertically below that of dorsal. About 50 scales in a longitudinal 
series ; ventral scutes not prominent. 

Several specimens, up to 170 mm. in total length, taken at Station 118, Port 
Stanley, Falkland Islands, in February 1904, when extraordinary shoals of this herring 
visited Port Stanley Harbour. 

Argentinid^. 

3. Bathylagus glacialis, sp. n. (PI. IX. fig. 2.) 

Depth of body 6 to 6^ in the length, length of head 4-^ to 4|-. Diameter of eye 
2i to 2^ in the length of head, interocular width 3, interorbital width 6. Dorsal 10 ; 
origin nearer to end of snout than to base of caudal. Anal 1 8. Pelvics 8-rayed, 
inserted below middle of dorsal. About 35 scales in a longitudinal series. 

* a series of nine water-colour drawings made by Mr Cuthbertson for the most part represent fishes from Scotia 
Bay, Soiith Orkneys, viz. Notolepis coatsii, Harpagifer bispitiis, Trematomus newnesii, Notothenia coriiceps, N. nudifrons, 
and N. gibberifrons ; there is also a sketch of Lycenchelys antarcticus. In one or two cases I have referred to these 
in the text. 



232 MR C. TATE REGAN ON THE 

There are five examples of this new species, 70 to 100 ram. in total length : — 

1. Station 398, 68° 25' S., 27° 10' W., 1 to 1000 fathoms; surface temperature 

30° F. ; vertical net ; 29th February 1904. 

2. Station 422; 68° 32' S., 12° 49' W., to 800 fathoms; surface temperature 

31-1° F. ; temperature at 800 fathoms 32-4° F. ; vertical net; 23rd March 
1904. 

3. Station 414, 71° 50' S., 23° 30' W., to 1000 fathoms; surface temperature 

29-1° F. ; vertical net; 15th March 1904. 

4. Station 417, 71° 22' S., 16° 34' W., 1410 fathoms; temperature at 1410 

fathoms 31-9° F. ; trawl; 18th March 1904. 

5. Station 418, 71° 32' S., 17° 15' W., 1221 fathoms; temperature at 1221 

fathoms 31-9° F. ; trawl ; 19th March 1904. 

Baihylagus antarcticus, Giinth., is distinguished by the less graceful form (depth 
5 in the length) and the longer anal fin with 22 rays. Bathylagiis gracilis, another 
Antarctic species recently described by Lonnberg, has the interorbital space very 
narrow and deeply concave, and about 41 scales in a longitudinal series. 

Other species of Bathylagus have been described from the South Atlantic (Gunther, 
Lonnberg), the North Atlantic (Goode and Bean), and the North Pacific (Gilbert). 

Galaxiid^. 

4. Galaxias attenuatus, Jenyns. 

Two examples from Port Stanley and Port Harriet, Falkland Islands, Station 118. 

5. Galaxias maculatus, Jenyns. 
Several from Port Harriet, Station 118. 

Haploghitonid^. 
6. Haplochiton zebra, Jenyns. 
One specimen from Port Stanley, Falkland Islands, in fresh water, Station 118. 

Stomiatidje. 

7. Stomias boa, Kisso. 

One from Station 451, 48° 06' S., 10° 5' W., 1742 fathoms; trawl; 13th April 

1904. 

8. Cyclothone microdon, Giinth. 

Small examples of this widely distributed species were taken at three stations, viz. — 
Six at Station 450, 48° 00' S., 9° 50' W., 1332 fathoms; surface temperature 

40-0° F. ; trawl; 13th April 1904. 
One at Station 422, 68° 32' S., 12° 49' W., 0-800 fathoms ; temperature at 800 

fathoms 32-4° F. ; vertical net; 23rd March 1904. 
Four at Station 414, 71° 50' S., 23° 30' W., 0-1000 fathoms; surface tempera- 
ture 29-1° F. ; vertical net ; 15th March 1904. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 233 

INIOMI. 

SUDID^. 

9. Notolepis coatsii, Dollo. 

Proc. Roy. Soc. Edin., xxviii., 1908, p, 58. 

Prymnothonus (part.), Glinth., ''Challenger" Pelagic Fish, p. 39, pi. v. fig. D (1889). 

., Hookeri (non Richards.), Dollo, Proc. Boy. Soc. Edin., xxvii., 1907, p. 35. 

Depth of body 6|- in the length, lengt*h of head 5 ; snout half the length of head ; 
diameter of eye 6^ in the length of head. Teeth rather small, pointed, uniserial, in 
jaws and on palatines. Dorsal 8 ; origin nearly equidistant from head and base of 
caudal ; adipose fin rather long and low. Anal 28. Caudal with numerous procurrent rays. 
Pectorals narrow, about ^ length of head. Vent below anterior part of dorsal. Scales 
deciduous. Myotomes 82, 34 in advance of dorsal fin. Silvery white ; back bluish. 

It is with some difficulty that I have put together the above description of the type of 
the species, 105 mm. in total length, taken at the surface in Scotia Bay, South Orkneys. 
The specimen is in very bad condition,* and everything that one touches falls off"; 
hence it is not surprising that I cannot see the small pelvic fins described by Dollo. 

In a paper on the classification of the Iniomi [Ann. Mag. Nat. Hist. (8), vii., 1911, 
pp. 120-133) I have already called attention to the fact that Dollo's family Pa.ralepidae 
is not a natural group, and that Notolepis diff"ers from Paralepis apparently only in 
the greater length of the adipose fin, a character of very slight importance to anyone 
familiar with the species of Siluroids. 

Larval and post-larval examples of this species that I have examined are : — 
1.— 44 mm. ; 62° 26' S., 95° 44' E. Challenger collection. 

2.— 50 mm. ; at Station 422, 68° 32' S., 12° 49' W., 10-800 fathoms; tempera- 
ture at 800 fathoms 32*4° F.; 23rd March 1904. Scotia collection. 
3-5.-38 to 56 mm. ; at Station 414, 71° 50' S., 33° 30' W., 0-1000 fathoms; 
surface temperature 29 '1° F. ; 15th March 1904. Scotia collection. 
Except that the teeth are relatively stronger and the eye larger, specimens 1 and 2 
are extremely similar to the type, and agree with it in the number of fin-rays and 
of myotomes ; I cannot find any pelvic fins, nor ascertain the position of the vent, but 
the eight-rayed dorsal fin is distinct in both. 

Specimens 3 to 5 are the ones described by Dollo as Prymnothonus hookeri ; these 
evidently belong to the same species as the other examples, with which the larger one 
agrees in the head, dentition, and approximate number of myotomes. In the smaller 
ones the head is relatively smaller and the snout shorter. I am unable to make out 
the fins, or position of vent, and I am very doubtful as to whether the so-called 
embryonic anal fringe is an actual structure present in the living fish. 

Dr Dollo named this species in honour of the late Mr James Coats, junr., of Paisley, 
whose generosity was the chief means of assuring the dispatch of the Scottish National 
Antarctic Expedition. 

* This is regrettable, as this specimen was originally so perfectly preserved and was brought home in perfect 
condition, and was acknowledged to have been received by Dr Dollo " eii hon e'tat." — W. S. B., Editor. 



234 



MR C. TATE REGAN ON THE 



Myctophid^. 

10. Myctophum antarcticum, Giinth. 
Specimens were taken at : — 

Station 309, 63° 51' S., 41° 50' W., 2300 fathoms; temperature 31 05° F. ; 

trawl; 16th March 1903. 
Station 414, 71° 50' S., 23° 30' W., 0-1000 fathoms; surface temperature 

29-1° F. ; vertical net ; 15th March 1904. 
Station 422, 68° 32' S., 12° 49' W., 0-800 fathoms; surface temperature 

327° F. ; temperature at 800 fathoms 32-4°; vertical net ; 23rd March 1904. 

1 1. Lampanyctus hrauey-i, Lonnberg. 

One specimen was taken at Station 420, lat. 69° 33' S., 15° 19' W., 2620 fathoms, 
by the trawl, on 21st March 1904 ; temperature 31'5° F. The species was previously 
known only from the type. 

Alepidosaurid^. 

Eugnathosaurus, gen. nov. 

Skull very elongate and strongly compressed, with the upper surface somewhat 
convex, bearing a fairly prominent median ridge. Snout and lower jaw much produced, 
each ending in a fleshy appendage ; lower jaw projecting beyond upper ; suspensorium 
directed obliquely forward. Teeth pointed, uniserial ; prsemaxillary teeth minute ; 
mandibulary teeth sub-conical, erect or somewhat retrorse, strongest in the middle of 
the length of the jaw, more spaced posteriorly ; palatine teeth strong, compressed, 
curved somewhat forward. 

12. Eugnathosaurus vorax, sp. n. 

The type of this remarkable new genus and species is a head, measuring 150 mm. 
in length from tip of snout to end of operculum, taken in the trawl on 18th March 1904, 




Fig. 1. — Eugnathosaurus vorax. 



at Station 417, in lat. 71° 22' S., long. 16° 34' W., off" Coats Land, at a depth of 1410 
fathoms, by the trawl; temp. 3 1*9° F. That it is related to Alepidosaunus is evident, 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 235 

but the form of the skull, the produced jaws, and the different mandibulary and palatine 
dentition distinguish it from that genus ; the antrorse palatine teeth are especially 
peculiar. 

The dentaries of a second specimen were taken at the same locality. 

APODES. 

Synaphobranchid^. 

13. Synaphohranchus australis, sp. n. (PI. VIII. fig. 5.) 
Synaphobranchus hathybius (part.), Giinth., " Ghallenger" Deep-Sea Fish, p. 254 (1887). 

The Challenger specimen, 350 mm. in total length, was taken midway between the 
Cape of Good Hope and Kerguelen, at a depth of 1375 fathoms, The Scotia example 
was obtained on 13th April 1904, at Station 451, in 48° 06' S., 10° 5' W., at a depth 
of 1742 fathoms, and measures a total length of 470 mm. The species belongs to 
the sub-genus Histiohranchus, Gill, which includes also S. hathyhiiis, Giinth., and 
S. infernalis, Gill. All three are closely related, diifering as follows : — 

Eye nearer to end of snout than to angle of mouth ; origin of dorsal above base of 
pectoral, its distance from end of snout rather less than ^ that from end of 
snout to vent .......... hathybius. 

Eye about equidistant from snout and angle of mouth ; origin of dorsal above 
posterior part of pectoral, its distance from end of snout somewhat more than 
\ that from end of snout to vent ...... infernalis. 

Eye about equidistant from snout and angle of mouth ; origin of dorsal a little 
behind end of pectoral, its distance from end of snout about 2^ in that from 
end of snout to vent ......... australis. 

ANACANTHINI. 

Macrurid^e. 

Four species of this family were obtained by the Scotia in Antarctic seas, all belong- 
ing to the sub-family Macrurinse (cf. Ann. Mag. Nat. Hist. (7), xi., 1903, pp. 459-466), 
and to genera with the teeth in the lower jaw uniserial. 

14. Nematonurus lecointei, Dollo. 

Res. Voy. " Belgica," Poiss., p. 44, pi. vii. (1904) ; Proc. Roy. Soc. Edin., xxix , 1909, p. 488. 

The type was taken in 70° 40' S., 102° 15' W., depth 1526 fathoms. The Scotia 
examples are from : (1) Station 313, 62° 10' S., 41° 20' W., 1775 fathoms; temperature 
31-0° F.; trawl; 18th March 1903. (2) Station 451, 48° 06' S., 10° 05' W., 1742 
fathoms; 13th April 1904.. 



236 MR C. TATE REGAN ON THE 

The praemaxillary teeth are biserial, except posteriorly, where the inner series is 
replaced by three, forming a narrow band. 



15. Chalinura ferrieri, sp. n. (PI. II. fig. 1.) 
Snout rather strongly produced (for a Chalinura) ; mouth wide, the maxillary 
nearly reaching the vertical from posterior edge of eye ; infraorbital ridge fairly 
prominent. Diameter of eye less than length of snout, 4f in length of head ; inter- 
orbital width 4. Dorsal 119; distance from second dorsal a little more than f the 
length of head. Origin of anal at distance from head equal to length of head without 
snout. Pectoral 18 or 19-rayed, f the length of head, extending to above origin of 
anal. Pelvics 11 -rayed, the outermost ray filamentous, reaching anal. Scales mostly 
with 3 parallel series of spinules, but the lateral series sometimes reduced to a single 
spine, or absent ; 8 scales between dorsal fin and lateral line. 

A single specimen, 230 mm. in total length, from Station 417, 71° 22' S., 16° 34' W., 
1410 fathoms, off" Coats Land; temperature at 1400 fathoms 3r9° F. ; trawl; 18th 
March 1904. 

This species is named after James G. Ferrier, Esq., F.R.S.Gr.S., Hon. Secretary of 
the Scotia Committee. 

16. Chaliyiura ivhitsoni, sp. n. (PI. II. fig. 2.) 
Snout rather produced (for a Chalinura) ; maxillary extending to below posterior 
margin of pupil ; infraorbital ridge prominent. Diameter of eye more than length of 
snout, 2f to 3^ in length of head ; interorbital width 4 to 4;^. Dorsal II 9-10 ; dis- 
tance from second dorsal \ the length of head. Origin of anal at distance from head 
equal to length of head without snout. Pectoral 18-19-rayed. Pelvic 9-rayed, the 
outermost ray filamentous, not reaching anal. Scales with 1 series of spinules, but 
some on sides of head with 3 series converging anteriorly ; 7 scales between dorsal fin 
and lateral line. 

Two specimens : — 

1. 420 mm.; Station 451, 48° 6' S., 10° 5' W., 1742 fathoms; trawl; 13th 

April 1904. 

2. 270 mm.; Station 417, 71° 22' S., 16° 34' W., 1410 fathoms, off" Coats Land; 

temperature at 1400 fathoms 31-9° F. ; trawl ; 18th March 1904. 

This species is named after T. B. Whitson, Esq., C.A., Hon. Accountant of the 
Scotia Committee. 

17. Cynomacrurus piriei, Dollo. (PL III. fig. 1.) 
Proc. Roy. Soc. Edin., xxix., 1909, p. 316. 
The type of the genus and species, a specimen of 300 mm., was obtained by the 
Scotia at Station .4 14, 7 1° 50' S., 23° 30' W., in a depth 0-1000 fathoms, surface tempera- 
ture 31*5° F., vertical net, on 15th February 1904. The dentition is very characteristic ; 
in the prsemaxillaries a narrow band of unequal teeth separated by an interspace from a 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 237 

marginal series of small teeth with a strong pair of antero- lateral canines ; in the lower 
jaw teeth strong, spaced, unequal, uniserial. 

Other important characters are the large terminal mouth with lateral cleft, the 
absence of a barbel, the small eye, and the slender, smooth dorsal spine. The pelvic fins 
are 7-rayed, and in counting 12 Dr Dollo must have reckoned divided rays as two. 

This species was named by Dr Dollo after Dr J. H. Harvey Pirie, bacteriologist, 
geologist, and surgeon of the Scotia. 

PERCOMORPHI. 

Serranid^. 
18. Csesioperca coatsii, sp. n. (PI. VI. fig. 1.) 

Depth of body 2f to 3| in the length, length of head 2f to 3. Diameter of eye 2^ 
to 2| in length of head, interorbital width 4^ to 5. Interorbital region flat ; maxillary 
extending to below middle of eye ; 21 to 24 gill-rakers on lower part of anterior arch. 
Dorsal X, 15-18; third or fourth spine longest, nearly twice as long as last, f to |- 
length of head. Anal III 8 ; second spine longest, as long as or longer than longest 
dorsal spine. Pectoral shorter than head, asymmetrical, the rays increasing to the 
tenth, counting from above, or seventh, from below. Caudal truncate. About 40 scales 
in a lateral longitudinal series and 45 in the lateral line, which forms an angle on the 
caudal peduncle. Pale reddish brown, with traces of alternating darker and paler longi- 
tudinal bands ; upper half of spinous dorsal blackish, or with a series of blackish spots. 

Gough Island. Several specimens, up to 135 mm. in total length, taken at Station 
461, 40° 20' S., 9° 56' W., off" Gough Island, at a depth of 100 fathoms; surface 
temperature 54-5° R; trawl; 23rd April 1904. 

This species is of considerable interest, as its three congeners are found on the coasts 
of South Australia, Tasmania, and New Zealand. These are distinguished by their 
longer pectoral and emarginate caudal fins, and by the convex interorbital region ; but 
I am unable to find any characters which would justify the establishment of a new 
genus for the new species. The pectoral fin of C. rasor is almost as asymmetrical, and 
I find that the flatness of the interorbital region of C. coatsii is not associated with any 
difference in the essential structure of the frontal bones, which are, as in C. lepidoptera, 
smooth and convex posteriorly, and anteriorly consist of a pair of supraorbital flanges 
and of a median depression bordered by muciferous canals. 

t have pleasure in naming this species after Major Andrew Coats, D.S.O., a 
member of the Scotia Committee, a most generous donor to the funds of the Scottish 
National Antarctic Expedition, and himself a polar explorer. 

Atherinid^. 
19. Basilichthys laticlavia, Cuv. and Val. 

Several small specimens from Station 118, Port Stanley Harbour, Falkland Islands, 

51° 41' S., 57° 51' W. ; shore. 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART IL (NO. 2). 31 



238 MR C. TATE REGAN ON THE 

ZOARCID^. 

20. Iluocoetes Jimhriatus, Jen5ms. 
Station 118, Port Stanley, Falkland Islands, 51° 41' S., 57° 51' W. ; shore. 

21. Austrolycus depressiceps, Regan. 

Several small specimens from Station 118, Port Stanley, Falkland Islands, 51° 41' S., 
57° 51' W. ; shore. This species is described on p. 245. 

22. Phucocoetes latitans, Jenyns. 

Four small specimens from Station 118, Port Stanley, Falkland Islands, 51° 41' S., 
57° 51' W. ; shore. 

23. Lycenchelys antarcticus, Regan. 

This new species is described on p. 242, from a single specimen from Station 313, 
62° 10' S., 41° 20' W.; depth 1775 fathoms; temperature 31-0° F. ; trawl; 18th 
March 1903. 

Brotulid^. 

24. Neohythites hrucei, Dollo. (PI. III. fig. 2.) 
Proc. Roy. Soc. Edin., xxvi., 1906, p. 172. 

Depth of body 6^ in the length, length of head 5, or If in its distance from origin 
of anal. Diameter of eye 13 in length of head, equal to width of posterior nostril; 
maxillary extending well behind eye ; palatine bands of teeth broad ; no prseopercular 
spines ; gill-membranes united for a short distance to each other and to isthmus ; 
15 gill-rakers on lower part of anterior arch. About 125 scales in a longitudinal series. 
Dorsal 108 ; origin behind base of pectoral. Anal 86. Pectoral nearly as long as 
head ; pelvics f the length of head, 2-rayed, each ray simple, expanded distally into an 
ovate blade. 

The type, 350 mm. in total length, was taken at Station 291, 67° 33' S., 36° 35' W. ; 
depth 2500 fathoms; trawl; 7th March 1903. 

From most species of Neohythites this species differs in the gill-membranes attached 
to the isthmus and the oar-shaped pelvic rays, and I should be inclined to recognise 
Garman's genus Holcomycteronus for this species and N. digittatus, had not Garman 
stated that the form of the pelvic rays is variable in the latter. 

This species was named by Dr Dollo after Dr W. S. Bruce, leader of the Scottish 
National Antarctic Expedition. 

BOVICHTHYID^. 

25. Cottoperca gohio, Giinth. 
Station 349, 51° 41' S., 57° 51' W., Port William, Falkland Islands; shore. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 239 

26. Cottoperca macrophthalma, Regan. 

At Station 346, 54° 25' S., 57° 32' W., Burdwood Bank; 56 fathoms; surface 
temperature 4r8° F. ; otter trawl; 1st December 1903. This new species is described 
on p. 253. 

27. Bovichthys diacanthus, Carmich. 

A specimen of 120 mm. from Gough Island. On comparison with Cliilian examples 
of the species usually known as B. diacanthus, I find that they are distinct {cf. p. 256). 

NOTOTHENIID^. 

28. Harpagifer hispinis, Forst. 
Several examples from Station 118, Porb Stanley, Falkland Islands, and Station 325, 
Scotia Bay, South Orkneys — the latter a new record of locality for this species. 

29. Trematomus newnesii, Bouleng. 
Station 325, Scotia Bay, South Orkneys. 

30. Trematomus horchgrevinkii, Bouleng. 
Station 325, Scotia Bay, South Orkneys. 

31. Trematomus hernacchii, Bouleng. 
Station 325, South Orkneys. 

32. Trematomus hansoni, Bouleng. 
Station 411, Coats Land, 161 fathoms. 

33. Notothenia tngram,ma, Regan. 

This new species, from Station 118, at the Falkland Islands, is described on p. 266. 

34. Notothenia ramsayi, Regan. 
This new species, from Station 346, the Burdwood Bank, is described on p. 267. 

35. Notothenia tesselata, Richards. 
Station 118, Port Stanley, Falkland Islands. 

36. Notothenia wiltoni, Regan. 

Examples of this new species, described on p. 268, were taken by the Scotia at 
Station 118, Port Stanley, Falklands, and at Station 346, the Burdwood Bank. 

37. Notothenia hrevicauda, Lonnb. 
Station 118, Port Stanley, Falkland Islands. 



240 MR 0. TATE REGAN ON THE 

38. Notothenia sima, Richards. 
Station 118, Port Stanley, Falkland Islands. 

39. Notothenia gihherifrons, Lonnb. 
Station 325, Scotia Bay, South Orkneys. 

40. Notothenia nudifrons, Lonnb. 
Station 325, Scotia Bay, South Orkneys. 

41. Notothenia coriiceps, Richards. 
South Orkneys ; common at Station 325, Scotia Bay. 

42. Notothenia cornucola, Eichards. 
Station 118, Port Stanley, Falkland Islands. 

43. Notothenia, rossi, Richards. 
Station 325, Scotia Bay, South Orkneys, 

44. Eleginops maclovinus, Cuv. and Val. 
Station 118, Port Stanley, Falkland Islands. 

Bathydraconid^. 
45. Bathydraco scotise, Dollo. 

Station 417, 71° 22' S., 16° 34' W., off Coats Land, at a depth of 1410 fathoms. 
This species is described on p. 282. 

SCLEROPAREl. 

SCORP^NIDAE. 

46. Sebastes maculatus, Cuv. and Val. 

Specimens from Station 461, Gough Island, at 25 fathoms and 100 fathoms, the 
latter witli Ciesioperca coatsi. 

47. Sebastes capensis, Gmel. 
A small specimen taken at Station 461, Gough Island, with the preceding; both 
these species are found at the Cape of Good Hope. 



ANTAHCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 241 

PSYCHROLUTID^. 

48. Neophrynichthys marmoratus, sp. n. 
Neophrynichthys latus (non Hutton), Giinth., Proc. Zool. Soc, 1881, p. 20, pi. i. 

In this species the dermal appendages on the head and anterior part of the body are 
much larger and set further apart than in N. latus. Another striking difference is the 
narrower interorbital region, its width measuring only ^ of the length of the head in 
N. marmoratus, but f in its congener from New Zealand. The dorsal rays number 
IX-X, 15-16, the anal 11 or 12 ; the caudal is more rounded than in N. latus. 

The irregular marbling gives this fish a very different appearance from the New 
Zealand form, with its definite pale spots separated by a brown network. 

Three specimens, two in the British Museum collection, from the Straits of 
Magellan, 320 and 390 mm. in total length, and one of 160 mm. obtained by the Scotia 
at Station 346, 54° 25' S., 57° 32' W., Burdwood Bank, 56 fathoms; surface 
temperature 41*8° F. ; otter trawl ; 1st December 1903. 

II. A Revision of the Zoarcid^ of Southern America and the Antarctic. 

The Zoarcidge are principally a northern family, and so far as I am aware none is 
known from South Africa, Australia, or New Zealand. Two northern deep-water genera, 
Lycenchelys and Melano stigma, are represented in the Antarctic Regions, but the littoral 
species, with those of South America and the adjacent islands, all belong to genera 
distinct from the northern ones."* There has hitherto been much confusion as to the 
characters of these genera and species, which it is the object of this revision to clear up. 

Synopsis of the Genera. 

I. Pelvic fins present ; mouth subterminal. 

A. Snout and lower jaw without fringes. 

1. Origin of dorsal fin well behind base of pectoral ; gill-opening cleft 

downward nearly or quite to lower end of base of pectoral. 
Teeth uniserial or biserial in jaws, uniserial on palatines ; tail long and 

slender . . . . . . . . .1. Lycenchelys. 

Teeth in jaws triserial ; two teeth near anterior end of each palatine ; tail 

moderately elongate ...... 2. Ophthalmolycus. 

2. Origin of dorsal fin above base or anterior part of pectoral ; gill- 

opening cleft downward at least to middle of base of pectoral. 
a. Mouth large, with wide lateral cleft ; gill-opening cleft down- 
wards almost or quite to lower end of base of pectoral ; 
teeth in jaws uniserial, with anterior canines in the upper 
and lateral canines in the lower. 

* The habitat of Gymnelis pidus, Giinth., is unknown, and there is no justification whatever for the statement 
that it comes from Magellan Straits. 



242 MR C. TATE REGAN ON THE 

Teeth on Tomer and palatines ...... 3. Iluoccntes. 

Palate toothless ....... 4. Lycodichthys. 

h. Mouth moderate, with short lateral cleft ; teeth in jaws 

uniserial laterally, usually bi- or tri-serial anteriorly ; no 

well marked canines ; teeth on palate. 

Head not depressed ; gill-opening cleft downward nearly to lower end of 

base of pectoral ...... 5. Austrolycichthys. 

Head depressed ; gill-opening cleft downward only to middle of base of 
pectoral ......... 6. Austrolycus. 

3. Origin of dorsal fin above base of pectoral ; gill-opening small, 
above the pectoral ; teeth in upper jaw uniserial, in lower bi- 
or tri-serial ; teeth on palate . . .7. Phucoccetes. 

B. Snout and lower jaw with dermal fringes ; palate toothless. 

Teeth conical, bi- or tri-serial ; gill-opening almost entirely above the 
pectoral . . . . . . . . .8. Crossolycus. 

Teeth incisor-like, uniserial ; gill-opening cleft downward to middle of 
base of pectoral . . . . . . . .9. Platea. 

11. No pelvic fins ; mouth terminal ; origin of dorsal just behind head ; teeth 
uniserial, in jaws and on vomer and palatines. 
Gill-opening cleft downward to middle of base of pectoral . 1 0. Maynea. 
Gill opening above base of pectoral . . . .11. Melanostigma. 

1. Lycenchelys, Gill, 1884. 

Proc. Acad. Philad., p. 110. 

Form elongate, with the tail long and slender ; mouth subterminal ; teeth in jaws 
slender, uni- or bi-serial ; teeth on vomer ; palatine teeth uniserial. Gill- opening rather 
wide, cleft downwards to lower end of base of pectoral. Dorsal origin well behind head ; 
pelvic fins present. 

Lycenchelys antarcticus, sp. n. (PI. IX. fig. 8.) 

Depth of body 16 in the length, length of head G and equal to its distance from 
origin of anal fin. Head as broad as deep, its breadth a little more than ^ its length. 
Snout twice as long as diameter of eye, which is 6 in length of head ; interorbital 
width about 16. Muciferous channels of sides of head and lower jaw with large pores. 
Lower jaw included ; teeth in jaws rather slender and obtuse, uniserial, biserial near 
symphysis of lower jaw; teeth on palate acute, wide-set. About 110 rays in dorsal 
fin, 9 in caudal, 103 in anal; origin of anal ^ as distant from vertical through origin 
of dorsal as from that through base of pectoral, which fin is a little more than ^ as long 
as head. Bluish grey ; head darker ; fins brownish grey. 

A single specimen, 128 mm. in total length, from near the South Orkneys, Station 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 243 

313, 62' 10' S., .41° 20' W., depth 1775 fathoms; bottom temperature 31-0° F. ; 
trawl; 18th March 1903. 

The few species of this genus hitherto described are from deep water north of the 
Equator. 

2. Ophthahnolycus, gen. nov. 

Form elongate, compressed. Mouth subterminal ; teeth rather slender and acute, 
in about 3 series in both jaws ; no canines ; 3 teeth on vomer and 2 near anterior end 
of each palatine. Gill-opening rather wide, cleft downwards nearly to lower end of 
base of pectoral. Dorsal origin well behind head ; pelvic fins present. 

Ophthalmolycus macrops. 
Lycodes macrops, Glinth., " Challenger" Shore Fish., p. 21, pi. xi. fig. B (1880). 

Depth of body 11^ in the length, length of head 5^. Diameter of eye 3^ in length 
of head and 7 times interorbital width. Maxillary nearly reaching vertical from 
posterior margin of eye. About 90 rays in the dorsal fin, 80 in the anal, and 10 in 
the caudal. Origin of dorsal above posterior ^ of pectoral ; origin of anal a head-length 
behind the head. Pectoral less than ^ the length of head. Yellowish ; 9 broad dark- 
brown cross-bars on back, extending on to dorsal fin ; a series of brown spots on the 
side, alternating with the bars ; a brown band from eye to operculum. 

Straits of Magellan, 40 to 140 fathoms. 

Here described from the type, 135 mm. in total length. 

Lycodes concolor, Roule {Bull. Mus. Paris, 1911, p. 280) may belong to this genus. 
D. 73 ; A. 68. Coloration uniform. 

3. Iluocoetes, Jenyns, 1842. 

Zool "Beagle," Fish., p. 165. 

Head about as broad as deep ; body compressed ; mouth subterminal, with wide 

lateral cleft ; teeth conical, uniserial in jaws, in a patch on the vomer and a single 

series on the palatines ; 1 or 2 pairs of canines at the symphysis of the upper jaw ; 

1 or 2 teeth on each side of lower jaw enlarged, canine-like. Gill-opening cleft 

downward to lower end of base of pectoral. Dorsal origin just behind head; pelvic 

fins present. 

Iluocoetes Jimbriatus. 

Jenyns, i.e., p. 166, pi. xxix. fig. 2. 

Lycodes variegatus, Giinth., Cat. Fish., iv. p. 322 (1868). 

Phucoccetes variegatus effusus, Smitt, Bihang Svensk. Vet.-AJcad., xxiv., 1898, iv.. No. 5, p. 43, 

pi. V. fig. 32. 
Phucoccetes variegatus micropus, Smitt, I.e., pi. v. fig. 33. 

Depth of body 8 to 11^ in the length, length of head 4|- to 5^. Diameter of eye 
4 to 5^ in length of head, 3 or 4 times the interorbital width. Maxillary extending to 
below posterior margin of eye. Dorsal with 80 to 85 rays, anal with 65 to 70, caudal 



244 MR C. TATE REGAN ON THE 

with about 10. Origin of dorsal above base of pectoral, of anal about a head-length 
behind head. Pectoral f to f the length of head. Head, body, and fins spotted and 
marbled ; sometimes cross-bars on the body ; a more or less distinct band from snout to 
eye and eye to operculum ; a series of blackish spots at margin of dorsal and anal. 

Falkland Islands ; Magellan Straits ; Chile. 

Here described from specimens from the Falkland Islands, 80 to 130 mm. in total 
length, including the types of Ly codes variegatus and two obtained by the Scotia at 
Station 118, Port Stanley. 

I am indebted to Mr L. Doncaster for the loan of Jenyns' type ; the appearance 
of some of the mucous canals as free fringing tubes is due to the bad state of preserva- 
tion of the specimen. 

Smitt's Phucoccetes variegatus elongafus {t.c, p. 44, pi. v. fig. 34) seems to be a 
distinct species, with the head § of the distance from operculum to origin of anal. 

4. Lycodichthys, Pappenheim, 1911. 

Sitzungsh. Gesellsch. Naturf. Freunde, 1911, p. 382. 

Closely related to Iluoccetes, differing especially in the toothless palate ; teeth 

uuiserial ; anterior pair in upper jaw enlarged ; lateral teeth of lower jaw spaced, 

canine-like. 

Lycodichthys antarcticus. 

Pappenheim, t.c, p. 383, and Deutsche Sudpolar-Exped., xiii., Zool., v. p. 180, pis. ix. fig. 6 and x. fig. 4. 

Depth of body 8 or 9 in the length ; length of head 5 to 5|. Diameter of eye 
5 to 6 in length of head. Maxillary extending to below posterior margin of eye or a 
little beyond. Dorsal with 85 to 90 rays, anal with about 65, caudal with about 10. 
Origin of dorsal a little behind base of pectoral, of anal 1 to 1^ head-lengths behind 
head. Pectoral }^ the length of head. Head, body, and fins spotted or marbled. 

Wilhelm Land. 

Here described from two co-types, 160 and 200 mm. in total length. 

5. Austrolycichthys, gen. nov. 
Closely related to Austrolycus, differing in the more compressed form, the head 
being at least as deep as broad, and in the more inferiorly placed and somewhat larger 
gill-openings, cleft downward nearly to the lower ends of the bases of the pectorals. 

(1) Austrolycichthys hrachycephalus. 

Lycodes hrachycephalus, Pappenheim, Deutsche Sudpolar-Exioed., xiii., Zool., v., p. 179, pi. x. fig. 3. 

Depth of body 8 to 10| in the length, length of head 5f to 6|. Tail from less than 
1| to If as long as rest of fish. Diameter of eye 5 in the length of head. Maxillary 
extending to below anterior part of eye. About 90 rays in the dorsal fin, 70 in the 
anal, 10 in the caudal. Origin of dorsal above anterior part of pectoral, of anal 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 245 

If to l| head-lengths behind head. Pectoral f or f the length of head. Grayish or 
brownish. 

Wilhelm Land, 380 metres. 

Here described from two co-types, 155 and 150 mm. in total length. These differ 
greatly in form and proportions, as is shown by the accompanying figures, but I cannot 
doubt that they belong to the same species. 





Fig. 2. — Austrolydclithyshrachycephalus. 

(2) Austrolycichthys bothriocephalus. 
Lycodes bothriocephalus, Pappenheim, Deutsche Sudpolar-Exped., xiii., Zool., v. p. 178, pi. x. fig. 2. 

Apparently related to the preceding species, the more slender of the two examples 
of A. hrachyceplialus described above showing considerable resemblance to the photo- 
graph of the type. But this species is said to have more numerous fin-rays, about 110 
in the dorsal and 90 in the anal. 

Wilhelm Land, 380 metres. 

Total length of the unique type, 181 mm. 

6. Austrolycus, gen. nov. 
Head depressed ; body compressed posteriorly. Mouth subterminal, with short 
lateral cleft ; teeth conical, uniserial on sides of jaws, bi- or tri-serial anteriorly ; teeth 
on vomer in a group, on palatines in a single series. Gill-opening cleft downwards to 
middle of base of pectoral. Dorsal origin just behind head ; pelvic fins present. 

(1) Austrolycus depressiceps, sp. n. (PL V. fig. 1.) 

Phucocoates latitans (nan Jenyns) Gixnth., Cat. Fish., iv. p. 321 (1862); Smith, Bihang Svensk. 
Vet.-Akad., xxiv., 1898, iv.. No. 5, p. 51, pi. v. figs. 37-39; Garman, Mern.. Mus. Camp. 
Zool., xxiv., 1899, p. 138. 

Depth of body about 10 in its length, length of head 5|- to 6 J. Diameter of eye 
6 1 to 9 in the length of head, much less than the interocular, but nearly equal to the 
TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 2). 32 



246 MR C. TATE REGAN ON THE 

interorbital width. Maxillaiy about reaching vertical from posterior margin of eye, 
100 to 110 rays in the dorsal fin, 70 to 80 in the anal, 8 to 10 in the caudal; origin 
of dorsal above base of pectoral, of anal If to 2 head-lengths behind head. Pectoral | 
the length of head. Brownish ; abdomen pale ; on side of head a sharp line between 
the dark brown above and pale yellow below, with the brown projecting downwards on 
the cheek as a bar ; a pale transverse band across nape, another above end of pectoral 
extending on to dorsal fin, which may be followed by similar bands or spots. 

Chile ; Patagonia ; Falkland Islands. 

Here described from a large series of specimens measuring up to 250 mm. in total 
length, including several obtained by the Scotia at Station 118, Port Stanley, 
Falkland Islands, 

(2) Austrolycus platei. 
Lycodes {Phucocoates) platei, Steind., Zool. Jahrb., Suppl. iv. p. 320, pi. xix. fig. 8 (1897-98). 

Evidently closely related to the preceding species, diff"ering in that the length of 
the head is | its distance from the vent, the tail is considerably longer than the head 
and trunk (only a little longer in A. depressiceps), and the coloration is diff"erent, the 
body being marked with broad cross bands, the interspaces between which correspond 
to the pale bands or spots on the back and dorsal fin of A. depressiceps. 

Chile, Cape Espiritu Santo. 

Total length 234 mm. 

This may be the Phucoccetes variegatus macropus of Smitt [Bihang Svensk. Vet.- 
Akad., xxiv., 1898, iv., No. 5, p. 44, pi. v. fig. 35). 



7. Phucoccetes, Jenyns, 1842.* 

ZooL ''Beagle;' Fish., p. 168 (1842). 

Head and body compressed. Mouth subterminal ; teeth conical, uniserial in upper 
jaw and on palatines, bi- or tri-serial in lower jaw ; anterior pair of teeth in upper jaw, 
middle vomerine tooth, and 1 or 2 pairs in lower jaw more or less enlarged and canine- 
like. Gill-opening small, above base of pectoral. Dorsal origin just behind head ; pelvic 
fins present. 

Phucoccetes latitans, Jenyns, I.e., pi. xxix. fig. 3. 

Lycodes flavus, Bouleng., Ami. Mag. Nat. Hist. (7), vi., 1900, p. 53. 

Depth of body 8 to 10 in the length, length of head 6^ to 7. Snout 1| as long as 
diameter of eye, which is 6 to 7 in length of liead, greater than interorbital width. 
Lower jaw included ; maxillary extending to below posterior part of eye. Dorsal with 

* Garman {Mem. Mus. Gomp. Zool, xxiv , 1899, p. 1.37) has described a fish from 16° N., 99° W., 660 fathoms, and 
has named it Phucoccetes suspectus. It is not a Phucoccetes, nor does it seem to be congeneric with any of the 
southern littoral forms. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 247 

about 100 rays, anal with about 80, caudal with 5 or 6. Origin of dorsal above base of 
pectoral ; pectoral f , pelvics \ as long as head. Brownish ; upper half of head dark 
brown, with a pale yellow band from eye to shoulder ; lower part of head pale yellowish. 

Falkland Islands. 

Here described from two specimens, 65 and 110 mm. in total length, the latter the 
type of L. Jlav us. Four small examples were obtained by the Scotia at Station 118, 
Port Stanley, Falkland Islands. 

8. Crossolycus, gen. nov. 

Form elongate, compressed. Snout and lower jaw with fringes. Mouth sub- 
terminal ; teeth in jaws conical, bi- or tri-serial ; lower jaw with a posterior canine ; 
palate toothless. Gill-opening almost entirely above base of pectoral. Dorsal origin 
above or a little in advance of base of pectoral ; pelvic fins present. 

(1) Crossolycus chilensis, sp. n. 
Lycodes (Iluocoetes) fimbriatus (non Jenyns) Steind., ZooL Jalirh., Suppl. iv., 1898, p. 322, pi. xx. fig. 10. 

Depth of body equal to length of head, 6f in the length of the fish. Diameter of 
eye 7 in length of head and equal to interorbital width. Lips thick. Dorsal 80. 
Anal 60. Distance from head to origin of anal 1^ the length of head. Pectoral f as 
long as head. Head, body, and dorsal fin marbled with brown. 

Chile, Cape Espiritu Santo. 

Steindachner's specimen measured 252 mm. 

(2) Crossolycus fasciatus. 
Iluocoetes fimbriatus swh-s,]). fasciatus, Lonnberg, Swedish S. Polar Exped., Fish., p. 20 (1905). 

Depth of body 7\ in the length, length of head 5. Diameter of eye 5f in the length 
of head and equal to interorbital width. Distance from head to origin of anal 1^ the 
length of head. Pectoral a little more than -| the length of head. Dark brown, with 
5 or 6 whitish transverse bars. 

Falkland Islands. 

Total length 74 mm. 

A specimen of 60 mm. recorded by Lonnberg from Tierra del Fuego, uniform yellow 
in colour and differing somewhat in proportions, may belong to another species. 

9. Platea, Steind., 1897. 

Zool. Jahrh., Suppl. iv. p. 323. 

Teeth in jaws uniserial, incisor-like ; palate toothless. Snout and lower jaw with 
fringes. Gill-opening cleft downwards to middle of base of pectoral. Dorsal origin 
above anterior part of pectoral. Pelvic fins present. 



248 MR C. TATE REGAN ON THE 

Platea insignis. 
Steind., I.e., pi. xx. tig. 12. 
Depth of body \A^ in the length, length of head 7f. Dorsal with about 100 rays, 
anal with about 90. Body with dark spots and bars. 
Chile, Cape Espiritu Santo. 
Total length 265 mm. 

10. Maynea, Cunningham, 1870. 

Trans. Linn. Soc, xxvii. p. 471. 

Gymnelichtliys, Fischer, Jahrh. Hamburg Wiss. Anst., ii., 1885, p. 60. 

Elongate, compressed. Mouth terminal ; teeth conical, uniserial, in jaws and on 
vomer and palatines. Gill-opening cleft downwards to middle of base of pectoral. 
Dorsal origin just behind head. No pelvic fins. 

(1) Maynea patagonica. 

Cunningham, I.e., p. 472 ; Giinth., Proc. Zool. Soc, 1881, p. 881, pi. ii. figs. C and D. 

Depth of body 10 or 11 in the length, length of head 6f to 7^. Diameter of eye 
5 to 6 in length of head ; interorbital region quite narrow. Maxillary extending to 
below anterior ^ or middle of eye. About 120 rays in dorsal fin, 95 in anal, 8 in 
caudal. Origin of dorsal above base of pectoral, of anal If to 1^ head-lengths behind 
head. Pectoral less than ^ as long as head. Yellowish, with broad brown cross- 
bars separated by narrower interspaces. 

Patagonia ; Falkland Islands. 

Here described from two specimens, the type from the Otter Islands, 150 mm. 
in total length, and an example of 90 mm. from the Magellan Straits. 

(2) Maynea antarctica. 

QymnelicMhys antardicus, Fischer, Jahrb. Hamburg Wiss. Anst., ii., 1885, p. 61, pi. ii. fig. 9. 

Maxillary extending to. below posterior maigin of eye. About 97 rays in the 
dorsal, 74 in the anal; origin of latter only 1^ head-lengths behind the head. No 
cross-bars. 

South Georgia. 

Total length 220 mm. 

11. Melanostigma, Giinth., 1881. 
Froc. Zool. Soc, p. 21. 

Compressed, elongate ; skin loose, smooth, naked. iMouth terminal, oblique ; teeth 
uniserial, in jaws and on vomer and palatines. Gill-opening small, above base of 
pectoral. Dorsal origin just behind head ; no pelvic fins. 

In addition to the species described below, this genus includes a few from deep 
waters of the North Atlantic and Pacific. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 249 

Melanostigma gelatinosum. 
Giinth., I.e. 
Depth of body about 10 in the length, length of head 6. Diameter of eye 3^ in 
lenoth of head, interorbital width about 12. Maxillary extending to below middle of 
eye. Distance from head to origin of anal equal to length of head ; pectoral nearly 
^ as long as head. Sides spotted and marbled with purplish grey ; end of tail blackish ; 
inside of mouth, gill-opening, and vent black. 
Magellan Straits, 24 fathoms. 
Here described from the type, a specimen of 140 mm. 

III. A Monograph of the Nototheniiformes. 

The division Nototheniiformes includes Percoids without pungent fin-spines, with 
the spinous dorsal, when developed, shorter than the long soft dorsal and anal, the 




Fir. 3.— Pectoral fin-skeleton of 1, Cottoperca gohio ; 2, Trematomus newnesii ; and 3, Notothenia coriiceps. 
cl, cleithrum ; sc, hypercoracoid (scapula) ; /, foramen ; cor, hyjjocoracoid ; m, metacoracoid process; 1, 2, 3, radials. 



principal caudal rays reduced in number (usually 14), the pectorals typically broad- 
based and the pelvics jugular, separated by an interspace, and each formed of a spine 
and 5 branched rays. There is a single nostril on each side. The structure and position 
of the pectoral radials is highly characteristic ; they are 3 in number, rather large flat 
plates ; all or 2 are inserted on the hypocoracoid, and the lowest is the narrowest and 
has its lower edge in contact with the metacoracoid process. In other osteological 
characters the more generalised types are very similar to the Perciformes. 



250 



MR C. TATE REGAN ON THE 



The group corresponds to the Nototheniidse of Boulenger and Dollo, with the 
addition of Pleuragramma, which does not at all resemble Leptoscojpus, and after the 
exclusion of Centropercis, evidently related to Champsodon, and of AcantJiaphritis 
(Pteropsaron), which is related to Hemeroccetes. Draconetta is allied to the 
Callionymidse, and its resemblances to Harpagifer are not due to affinity. As now 
restricted the Nototheniiformes are characteristic of and peculiar to the Antarctic seas 




"dough Is- 



o Bou\/etk. 
S.aeor^ c» '< S.Saodu,.ch Group 



fS. Orkneys 







e Marion Is. 

»• Crozetls. 



Fig. 4. — Map showing the localities where Nototheniiform Fishes have been collected. 

and the region immediately to the north, ranging to S.E. Australia, New Zealand, 
Chile, Argentina, Tristan da Cunha, and St Paul Island. 

There is every reason to suppose that the group has always been an Antarctic one, 
and seeing that it has become differentiated into four quite distinct families and into 
several genera, we may perhaps infer that there has been a large cold southern ocean 
throughout the greater part of the Tertiary period. 

The group throws no light on the question of former extensions northward of the 
Antarctic Continent ; at the present day there are littoral species common to Australia 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 251 

and New Zealand [Bovichthys variegatiis), to New Zealand and South America 
[Notothenia macrocephala, N. cornucola), or to the Antarctic Continent and Kerguelen 
{Notothenia coriiceps, Harpagifer hispinis) ; and if under the existing conditions 
species may have this wide distribution, the fact that some are more restricted and are 
separated from the most nearly related forms by wide expanses of ocean can be 
explained without the theory of land-bridges. 

Many of the more southern types appear to be circumpolar ; for example, 
Trematomus newnesii, T. horchgrevinkii, T. bernacchii, T. hansoni, T. loennhergii, 
Plenragramma antarcticum, Notothenia coriiceps. 

With the exception of Pseudaphritis urvillii from the rivers of Tasmania and S.E. 
Australia the Nototheniiform fishes are marine, and the great majority of them are 
littoral ; several have been described as frequenting the rocks and weeds, but others 
prefer deeper water, the species varying in this respect like the Cottids and Gobies of 
our northern seas. Fishes pertaining to four genera (Bathydraco, Gerlachea, 
Racovitzaia, Cryodraco) live in the -open sea, and probably at some distance below 
the surface. 

Most of the fishes of this group feed on crustaceans, molluscs, etc. {cf. Lonnberg, 
Fish. Swedish South Polar Exped., p. 55), but Gymnodraco and the Chaenichthyidse 
are no doubt piscivorous. According to Lonnberg {I.e., p. 52) the breeding season varies, 
some species probably spawning in the spring, others in the summer, others in the 
autumn. The eggs are smaller in Notothenia and Trematomus than in Artedidraco 
and Champsocephcdus ; they are probably demersal in all, but certainly in the last two 
genera. 

Synopsis of the Families. 

I. One radial on hypercoracoid, two on hypocoracoid ; gill-membranes separate, 

free from isthmus ; teeth on vomer and palatines ; mouth protractile ; snout 
not produced ; a spinous dorsal fin . . . . 1. BovichthyidsB. 

II. All three radials on hypocoracoid ; gill-membranes united, free or attached to 

isthmus, usually forming a fold across it ; palate toothless. 

A. Palatine and pterygoids normally developed. 

Mouth protractile: snout not produced; a spinous dorsal fin 2. Nototheniidse. 
Mouth not protractile ; snout produced ; no spinous dorsal fin. 

3. BathydraconidsB. 

B. Palatine in great part ligamentous ; no mesopterygoid ; mouth not protractile ; 

snout produced and depressed .... 4. Chsenichthyidse. 

Family 1. BoviCHTHYiDyE. 

This family includes Nototheniiformes more generalised than the rest in the presence 
of bands of cardiform or villiform teeth not only in the jaws, but on the vomer and 
palatines, and in the separate free gill-membranes. All other members of the group 



252 MR C. TATE REGAN ON THE 

have the palate toothless and the gill-membranes united, or joined to the isthmus. 
The snout is not produced, the mouth is protractile, the lateral line is complete and 
continuous, and a spinous dorsal fin is present. The skeleton is well ossified ; there are 
2 radiais on the hypocoracoid and 1 on the hypercoracoid (fig. 3, 1) ; the palatine and 
pterygoids are normally developed. The vertebrae number 38 to 42 (13-16 + 22-29) ; 
prsecaudals with parapophyses from the fifth or sixth ; ribs and epipleurals on parapo- 
physes, when these are developed. 

Littoral fishes, with one species in fresh water. 

Three genera. 

1. Fseudaphritis, Casteln., 1872. 
Proc. Zool. Soc. Victoria, i. p. 72; Ogilby, Proc. Linn. Soc. N.S. Wales, xxii., 1897, p. 559. 

Body subcylindrical, fully scaled. Head small, scaly, somewhat depressed, narrowed 
forward ; interorbital region flat. Teeth in bands in jaws and on vomer and palatines ; 
lower jaw projecting. Operculum normal, with a weak spine ; gill-membranes not 
united, free from isthmus. Origin of spinous dorsal at some distance behind head ; rays 
of all the fins branched. 

S.B. Australia and Tasmania ; fresh- water. 



Pseudaphritis urvillii. 

Aphritis urvillii, Cuv. and Val., Hist. Nat. Poiss., viii. p. 484, pi. 243 (1831): Giinth., Oat. 

Fish., ii. p. 242 (1860). 
Pseudaphritis bassii, Casteln., I.e. 

Depth of body 5 to 6 in the length, length of head 3| to 4. Diameter of eye 
51 to 71 in the length of head, interorbital width 8 to 12. Maxillary extending to 
below eye ; about 10 short gill-rakers on lower part of anterior arch. Dorsal VII- VII I, 
19-20. Anal 23-25. 60 to 65 scales in the lateral line. Body marbled ; dorsal and 
caudal spotted. 

Rivers from New South Wales to South Australia and Tasmania. 

Here described from eight specimens, 100 to 240 mm. in total length, from South 
Australia, Victoria, and Tasmania. 



2. Cottoperca, Steind., 1876. 

Sitzungsb. Akad. Wien, Ixxii. p. 66. 

Head and body compressed, fully scaled. Head large ; snout broad ; interorbital 
region concave. Teeth in bands in jaws and on vomer and palatines ; lower jaw some- 
what the shorter. Operculum normal, with a weak spine ; gill-membranes not united, 
free from isthmus. Spinous dorsal originating above operculum ; rays of soft dorsal and 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 253 

anal unbranched ; lower pectoral rays simple, more or less thickened and partly 
free distally. 

Patagonia ; Magellan Straits ; Falkland Islands. 

(1) Cottoperca gohio. (PI. IV. fig. 3.) 

Aphritis gohio, Giinth., Ann. May. Nat. Hist. (3), vii., 1861, p. 88; and ^'■Challenger" Shore 

Fish., p. 21, pi. ix. (1880). 
Cottoperca rosenhergii, Steind., Sitzungsh. Ahad. Wien, Ixxii., 1876, p. 67, pi. v. fig. 1. 

Depth of body about 4 in the length, length of head about 2 J. Diameter of eye 
4 to 8 in the length of head, interorbital width 13 to 16. Maxillary extending to 
below posterior part or posterior edge of eye ; 5 or 6 short gill-rakers on lower part of 
anterior arch. Dorsal (VI) VII, 22-23. Anal 20-23. Dorsal spines and rays increasing 
in length with age, the longest varying from 3 to | the length of head. Pectoral 
about ^ the length of head ; six lowest rays simple, somewhat thickened. Caudal 
subtruncate. Least depth of caudal peduncle greater than the diameter of eye, except 
in quite young specimens. About 60 scales in a lateral longitudinal series, or 65 in the 
lateral line, which is complete and continuous. Orange-yellow, with three broad 
brownish cross-bars on upper part of body ; head and sides of body spotted and marbled 
with brown. 

Patagonia ; Tierra del Fuego ; Falkland Islands. 

Here described from nine specimens, 130 to 480 mm. in total length, from Magellan 
and the Falklands, at depths varying from 6 to 147 fathoms, including the types of the 
species and a specimen from Station 349, Port William, Falkland Islands, taken by the 
Scotia in January 1903. 

(2) Cottoperca macrophthalma, sp. n. (Pis. IV. fig. 2, and V. fig. 2.) 
Depth of body 4 to 5 in the length, length of head (to opercular spine) 2f to 2f , 
Diameter of eye 3^ to 5 in the length of head, interorbital width 13 to 16. Maxillary 
extending to below posterior part or margin of eye, or a little beyond. 5 to 7 short 
gill-rakers on lower part of anterior arch. Dorsal VII (VIII), 21-24. Anal 20-22. 
In the young, first dorsal spine longest, ^ the length of head and as long as soft rays ; 
in the adult, fourth or fifth spine longest, sometimes ^ the length of head ; longest soft 
rays sometimes f the length of head. Other fins, scales, coloration, etc., as in C. gobio. 
Least depth of caudal peduncle not more than diameter of eye. 

Ten specimens from Station 346, the Burdwood Bank, south of the Falkland Islands, 
54° 25' S., 57° 32' W., taken by the Scotia in 56 fathoms on 1st December 1903, and 
three from Magellan Straits, 100 to 450 mm. in total length. 

(3) Cottoperca macrocephala. 
Roule, Bull. Mus. Paris, 1911, p. 277 (1912). 
Eye large. Head longer and fins lower than in C. macrophthalma. Seven simple 
pectoral rays. Patagonia. 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART 11. (NO. 2). 33 



254 



MR C. TATE REGAN ON THE 



3. Bovichthys, Cuv. and Val., 1831. 
Hist, Nat. Poiss,, viii. p. 486. 

Differs from Cottoperca in the naked head and body, the abnormal operculum, 
which has a very strong spine and a superior process which articulates with the post- 
temporal, and in the enlarged and partly free posterior anal rays. 




Fig. 5. — Distribution of Bovichthys and Trematomus, respectively the iiortliernmost and southernmost Nototheniiform genera. 

B — Bovichthys. 1. B. variegatus ; 2. B. angustifrons ; 3. B. diacanthus ; 4. B. chilcnsis ; 4a. B. sp. ; 5. B. veneris; 

6. B. dccijncns ; 7. B. j^srjchrolutes ; 8. B. roseopictus. 

T = 2'reviaiomus. 1. T.nevmesii; 2. T. nicolai ; 3. T. borchgrcvinkii ; 4. T. br achy soma ; 5. T.vicariiis ; 6. T. bernacchii ; 

7. T.hansoni; 8. T. loennbergii. 

Note that in Bovichthys none, in I'rematomus most, of tlie species appear to be circumpolar. 



Chile ; Argentina ; Tristan da Cunha and Gough Island ; St Paul ; S.E. Australia 
and Tasmania ; New Zealand and neighbouring islands. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 255 



uj 



Synopsis of the Species. 

[. Interorbital region distinctly concave. 

Interorbital width -jV, pectoral fin | the length of head . 1. variegatus. 

Interorbital width -^ or -^^, pectoral fin f the length of head . 2. angustifrons. 
Interorbital width -^, pectoral fin -f the length of head . 3. diacanthus. 

II. Interorbital region nearly flat. 

Interorbital width \ the length of head, pectoral fin f . . .4. chilensis. 
Interorbital width more than | the diameter of eye, which is ^ the length of 

head ; pectoral extending to vent ...... 5. veneris. 

Interorbital width 6| in length of head ; pectoral extending well beyond origin 

of anal .......... 6. decipiens. 

Interorbital width 5 in length of head ..... 7. psychrolutes. 

Interorbital width 3^ in length of head . . . . §. roseopictus. 

(l) Bovichthys variegatus. 

Richards., ''Erebus" and ''Terror" Fish., p. 56, pi. xxxiv. figs. 1-4 (1846); Giinth., Gat. Fish., 
ii. p. 250 (1860). 

Depth of body 4 to 5 in the length, length of head 2f to 3. Diameter of eye 4 to 
4| in the length of head, interorbital width 10. Interorbital region moderately 
concave ; maxillary extending to below anterior ^ of eye ; opercular spine equal to or 
less than diameter of eye ; 8 gill-rakers on lower part of anterior arch. Dorsal VII- 
VIII, 18-20. Anal 14-15. Pectoral | the length of head. Caudal subtruncate. 
Caudal peduncle longer than deep. Body with irregular dark cross-bars and usually 
with pale spots and vermiculations ; spinous dorsal marbled, sometimes with a blackish 
blotch posteriorly ; soft dorsal, caudal, and pectoral with series of dark spots ; anal 
with a dark longitudinal band. 

S.E. Australia ; New Zealand and neighbouring islands. 

Here described from six specimens, 75 to 200 mm. in total length, including the 
types of the species, from New South Wales (Haslar), New Zealand [Otago Mus., 
Hutton), and Campbell Island {Southern Cross). 

(2) Bovichthys angustifrons, sp. n. (PL IV. fig. 1.) 
Depth of body 4| to 5 in the length, length of head 2f to 2|. Diameter of eye 
4 in the length of head, interorbital width 12 or 13. Interorbital region moderately 
concave ; maxillary extending to below anterior -^ of eye ; opercular spine nearly as 
long as diameter of eye ; 8 gill-rakers on lower part of anterior arch. Dorsal VIII, 19. 
Anal 14. Pectoral | the length of head. Caudal subtruncate. Caudal peduncle longer 
than deep. Dark spots on the head and blotches or bars on the body ; soft dorsal and 
caudal with series of spots on the rays. 

Here described from two specimens, 160 and 145 mm. in total length, the former 
from Tasmania (Allport), the latter without locality {Chatham Museum). 



256 MR C. TATE REGAN ON THE 

(3) Bovichihys diacanthus. (PI. IX. fig. 5.) 

Gallionymus diacanthus, Carmich., Trans. Linn. Soc, xii., 1818, p. 501, pi. xxvi. 
Bovichihys diacanthus, Giinth., Cat. Fish., ii. p. 249 (1860). 

Depth of body 5 in the length, length of head 3. Diameter of eye i^ in the 
length of head, interorbital width 11. Interorbital region concave ; maxillary extend- 
ing to below anterior ^ of eye ; opercular spine f the diameter of eye ; 9 gill-rakers on 
lower part of anterior arch. Dorsal VIII, 20. Anal 15. Pectoral -| the length of 
head. Caudal subtruncate. Caudal peduncle longer than deep. 

Tristan da Cunha ; Gough Island. 

Here described from a specimen of 120 mm. obtained by the Scotia at Gough 
Island, shore. 

The species was originally described from Tristan da Cunha, where it is said to be 
very common among the rocks and to attain a length of 7 inches. Carmichael 
describes the colour as olive, with green blotches and white dots. 

(4) Bovichthys chilensis, sp. n. 

Bovichihys diacanthus (non Carmichael), Cuv. and Val., Hist. Nat. Poiss., viii. p. 487, pi. 244 
(1831); Steind., Zool Jahrh., Suppl. iv., 1897-8, p. 300, pi. xx. fig. 1. 

Depth of body 5 in the length, length of head 3 to 3^. Diameter of eye 4^ to 4| 
in the length of head, interorbital width 9. Interorbital space nearly flat ; maxillary 
extending to below anterior \ of eye ; opercular spine as long as or shorter than eye ; 
8 gill-rakers on lower part of anterior arch. Dorsal VIII, 21. Anal 14-16. Pectoral 
f the length of head. Caudal truncate. Caudal peduncle longer than deep. Body 
marbled ; spinous dorsal dusky, with a dark blotch posteriorly ; soft dorsal with 2 or 3 
series of dark spots ; caudal dusky with orange posterior margin ; lower fins orange, 
more or less spotted. 

Chile ; Juan Fernandez. 

Here described from two specimens from Chile {DelJIn), 92 and 96 mm. in total 
length. 

Berg {Ann. Mus. Buenos Aires, ii., 1897, p. 298) has recorded a species of 
Bovichthys from Argentina as Boviclithys diacanthus. This may prove to be B. 
chilensis ; more probably it is a new species, as yet undescribed. 

(5) Bovichthys veneris. 

Bovichthys psychrolutes {non Giinth.), Kner, Novara Fische, p. 128, ph vi. (1869). 
„ veneris, Sauvage, Arch. Zool. Exp., viii., 1879, p. 25. 

Depth of body 5 in the length, length of head about 3 1. Diameter of eye 4 in the 
length of head. Interorbital region only slightly concave, its width rather more than 
I the diameter of eye. Maxillary extending to below anterior \ of eye ; opercular 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 257 

spine I the diameter of eye. Dorsal VIII, 19-20. Anal 14-15. Pectoral f or -| the 
length of head. Caudal rounded or subtruncate. Caudal peduncle longer than deep. 

Island of St Paul. 

Kner's and Sauvage's descriptions are based on specimens of 9 or 10 inches. 

(6) Bovichthys decipiens, sp. n. (PI. IX. fig. 1.) 

Depth of body 4 in the length, length of head 2f. Diameter of eye 3| in the 
length of head, interorbital width 6 J. Interorbital space nearly flat ; maxillary 
extending to below anterior ^ of eye ; opercular spine a little shorter than eye ; 8 gill- 
rakers on lower part of anterior arch. Dorsal VIII, 19. Anal 14. Pectoral f the 
length of head. Caudal peduncle longer than deep. Body with cross-bars ; spinous 
dorsal with a blotch posteriorly ; dorsal and pectoral with series of spots ; caudal barred. 

A specimen of 41 mm. from Cook's Straits {Hector) is very similar in appearance to 
B. variegatus. 

(7) Bovichthys psychrolutes. 
Giinth., Cat. Fish., ii. p. 250 (1860). 

Depth of body 4 in the length, length of head 3. Diameter of eye 3i in the length 
of head, interorbital width 5. Interorbital region nearly fiat ; maxillary extending to 
below anterior ^ of eye ; opercular spine as long as eye ; 8 gill-rakers on lower part of 
anterior arch. Dorsal VIII, 20. Anal 14. Pectoral f the length of head. Caudal 
subtruncate. Caudal peduncle longer than deep. Bluish-olive ; fins pale. 

Here described from the type, a specimen 38 mm. in total length, from S.W. of 
the Antipodes Islands (50° S., 170° W.). 

(8) Bovichthys roseopictus. 
Hutton, Trans. N. Zeal. Inst., xxxvi., 1903, art. ix. 

Depth of body 5|, length of head 4| in total length. Diameter of eye 3 in length 
of head, interorbital width 3^. Top of head smooth, with two small ridges. Dorsal 
VIII, 18 (?). Anal 13 (?). Pectoral as long as head. Caudal apparently truncated. 
Back dark olivaceous brown, sides and abdomen silvery ; a pink spot at base of operculum 
and 5 bright rose-pink bands on each side. 

New Zealand ; known from a single specimen of 46 mm. picked up on the beach at 
Sumner, Canterbury. 

Family 2. Nototheniid/E. 

Diff"er from the Bovichthyidse in the toothless palate, the united gill-membranes, 
and in having all 3 radials on the hypocoracoid (fig. 3, 2 and 3). Vertebrae 45-56 
(16-20 -F 25-35). 

In the typical genera the skeleton is well ossified, and the rather strong ribs and 
epipleurals are inserted on well-developed parapophyses, or only the first one or two 
are sessile. In Pleuragramma the skeleton is weak, with the bones thin and papery. 



258 MR C. TATE REGAN ON THE 

the vertebral centra are thin cylinders of bone, parapophyses are developed on the 
posterior prsecaudals only, and the ribs and epipleurals are feeble. 

Synopsis of the Genera. 

I. Body scaly ; gill-membranes forming a fold across the isthmus ; opercles normal. 

A. Hypercoracoid enclosing its foramen. 

Lateral line scales with tubules or pits ; . . 1. Trematomus. 

Lateral line scales merely notched. . . .2. Pleuragramma. 

B. Foramen partly bordered by hypocoracoid. 

1. Two or three lateral lines ; maxillary usually extending to below eye ; 

pectoral rounded or vertically truncated. 
Teeth in bands . . . . . . . . .3. Notothenia. 

Teeth uniserial . . . . . . . . 4. Dissostichus. 

2. One lateral line ; maxillary not reaching eye in the adult fish ; pectoral 

very obliquely truncated, the upper rays longest . 5. Eleginops. 

n. Body naked ; gill-membranes broadly united to isthmus ; operculum hooked 

upwards posteriorly, its upper edge deeply concave ; foramen partly bordered 

by hypocoracoid. . 

A mental barbel ; opercles not spinate . . . . . 6. Artedidraco. 

No barbel ; operculum and suboperculum each forming a strong spine. 

7. Harpagifer. 

1. Trematomus, Bouleng., 1902. 
"Southern Cross" Pisces, p. 177. 

Body scaly ; 2 lateral lines with tubular or pitted scales. Mouth moderate or 
rather large ; jaws with bands of villiform teeth. Gill-membranes united, free or 
forming a free fold across isthmus. Skeleton well ossified; vertebrae 52-56 (17-21 
-1- 32-35) ; most of the prseeaudals with parapophyses to which the ribs and epipleurals 
are attached ; hypercoracoid enclosing its foramen (fig. 3, 2). A spinous dorsal fin ; 
pectoral rounded or sub-vertically truncated. 

Coasts of the Antarctic Continent ; South Orkneys and South Georgia (fig. 5, p. 254). 

The difference between Notothenia, with the hypocoracoid bordering the foramen, 
and Trematomus, with the foramen enclosed in the hypercoracoid, may not be very 
important, and Pappenheim believed that he found both conditions in one species 
[Deutsche Sildpolar-Exped., xiii., Zool., v. p. 166, figs.); but this seems to have been 
an error, the specimen with perforate hypercoracoid being Trematomus hansom and not 
Notothenia lejndorhinus. Pappenhkim [t.c., p. 170) states that T. hernacchii is a 
Notothenia in the structure of its pectoral arch ; I have examined a large series of 
specimens, and find that the hypercoracoid encloses its foramen in all. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 259 

Synopsis of the Species. 

I. Upper surface of head naked. 

A. Cheeks and opercles fully scaled. 

Interorbital width 3^ to 5 in length of head. D VI-VIII, 32-38. 
A 32-36 ........ 1. newnesii. 

Interorbital width 8 or 9 in length of head. D IV, 37. A 32-33. 

2. nicolai. 

B. Cheeks and opercles scaly above, naked below ; interorbital width 3 to 5| 

in length of head. 

D V-VI, 34-37. A 31-33 . . .3. horchgrevinkii. 

D IV-V, 30-33. A 29-30 .... 4. hrachysoma. 

II. Occiput scaly. 

A. Interorbital region naked, or not fully scaled. D IV-VI, 33-38. A 31-35. 

Diameter of eye 4f in length of head, interorbital width about 5 
(in a specimen of about 270 mm.) ; 55 to 59 scales in a longi- 
tudinal series ........ 5. vicarius. 

Diameter of eye 3 to 4^ in length of head, interorbital width 5 to 9 
(in specimens up to 340 mm.); 60 to 75 scales in a longitudinal 
series ........ 6. hernacchii. 

B. Interorbital region fully scaled ; width 5 to 1 in length of head. 

D V-VII, 36-41. A 33-36 ... . . .7. hansoni. 

D V-VI, 33-34. A 31-33 .... 8. loennhergii. 

(l) Trematomus newnesii. 

Bouleng., "Southern Gross" Pisces, p. 177, pi. xi. (1902). 

Notothenia cyaneobrancha, VailL, Exped. Antarct. Fran^aise, Poiss., p: 26 (1906). 

„ microlepidota, Vaill., t.c, p. 35. 

„ hodgsoni, Bouleng., Nat. Antarctic Exped., Nat. Hist., ii., Fish., p. 2, pi. i. fig. 2 (1907). 

Depth of body 4 to 5 J in the length, length of head 3^ to 4|^. Diameter of eye 3 
to 4^ in the length of head, interorbital width 3-^ to 5. Maxillary extending to below 
anterior part or middle of eye (young) or beyond (adult) ; upper surface of head naked, 
cheeks and opercles scaly ; 15 to 20 gill-rakers on lower part of anterior arch. Dorsal 
VI-VIII, 32-38. Anal 32-36. Pectoral f the length of head or more, longer than 
pelvics, which reach the vent in the young, but not in the adult. Caudal truncate. 
Caudal peduncle about as long as deep. 68 to 86 scales in a longitudinal series from 
above base of pectoral fin to caudal, 40 to 52 in upper lateral line, which ends below 
posterior rays of dorsal, 3 to 19 in lower lateral line. Brownish, usually spotted or 
marbled or with irregular cross-bars ; spinous dorsal blackish ; other fins dusky, often 
with small dark spots. 

Here described from a large series of specimens, 50 to 200 mm. in total length, in- 
cluding the types of the species and of N. hodgsoni. The types came from Duke of 



260 MR C. TATE REGAN ON THE 

York Island, near Cape Adare, 3 to 5 fathoms, and Cape Adare, 4 to 8 fathoms ; those of 
N. hodgsoni from the Discovery winter quarters, Eoss Island. The Scotia specimens 
are all from Station 325, Scotia Bay, South Orkneys ; others in the British Museum 
are from the South Shetlands. 

(2) Trematomus nicolai. 
Notothenia nicolai, Bouleng., '■'Southern Gross" Pisces, p. 184, pi. xv. (1902). 

Depth of body nearly 4 in the length, length of head 3^ to 3|. Diameter of eye 
3 to 3^ in the length of head, interorbital width 8 to 9. Maxillary extending to below 
anterior ^ or ^ of eye ; upper surface of head naked ; cheeks and opercles scaly ; 
11 or 12 gill-rakers on lower part of anterior arch. Dorsal IV, 37. Anal 32-33. 
Pectoral | to f length of head, somewhat longer than pelvics, which reach vent or origin 
of anal. Caudal rounded. Caudal peduncle nearly as long as deep. 58 to 62 scales 
in a longitudinal series from above base of pectoral fin to caudal, 39 to 43 in upper 
lateral line, which ends below posterior rays of dorsal, 8 to 18 in lower lateral line. 
Brownish, with dark cross-bars and sometimes with small dark spots ; fins dusky. 
Victoria Land. 

Here described from the types, three specimens 150 to 250 mm. in total length, 
from Cape Adare, 5 to 8 fathoms, and Duke of York Island, near Cape Adare, 4 fathoms. 

The pectoral arch of this species is exactly similar to that of the closely related 
T. newnesii, as figured on p. 249. 

(3) TrcTnatornus borchgrevinkii. 

Bouleng., "Southern Cross" Pisces, p. 179, pi, xii. (1902); Pappenheim, Deutsche Siidpolar- 
Exped., xiii., Zool, v. p. 171 (1912). 

Depth of body 4 to 5 in the length, length of head 3| to 4 J. Diameter of eye 4 to 5 in 
length of head, interorbital width 3 to 4. Maxillary extending to below anterior ^ of 
eye ; upper surface of head naked ; upper parts of cheeks and opercles scaly ; 16 to 19 
gill-rakers on lower part of anterior arch. Dorsal V-VI, 34-37. Anal 31-33. 
Pectoral f to |^ the length of head, longer than pelvics, which rarely reach the vent. 
Caudal rounded or subtruncate. Caudal peduncle as long as deep, or deeper than long. 
78 to 96 scales in a longitudinal series from above base of pectoral fin to caudal ; lateral 
lines vestigial, without or with only a few tubules. Yellowish, with dark spots or 
irregular cross-bars ; dorsal and caudal sometimes with series of spots. 

Graham Land and neighbouring islands ; Wilhelm Land ; Victoria Land. 

Here described from several specimens, 180 to 270 mm. in total length, including 
the types of the species, from Cape Adare and Duke of York Island, near Cape Adare 
[Southern Cross), and examples from the Discovery winter quarters, E,oss Island. A 
specimen of 80 mm. was obtained in March 1903 by the Scotia, at Station 325, in 
Scotia Bay, South Orkneys; depth 10 to 15 fathoms. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 261 



(4) Trematomus hrachysoma. 
Pappenheim, Deutsche Siid'polar-Exped., xiii., Zool., v. p. 172 (1912). 

Depth of body i^^ to 4|- in the length, length of head 3 to 3f . Diameter of eye 
32 to 3| in the length of head, interorbital width 4 to 5^. Maxillary extending to , 
below anterior part or middle of eye ; upper surface of head naked ; upper parts of 
cheeks and opercles scaly ; 16 to 19 gill-rakers on lower part of anterior arch. Dorsal 
IV-V, 30-33. Anal 26-30. Caudal rounded. Caudal peduncle a little deeper than 
long. 65 to 75 scales in a longitudinal series. Yellowish brown ; head and back dark ; 
a series of 6 dark spots from operculum to caudal, and 5 below them at level of base of 
pectoral ; a dark spot at tip of spinous dorsal ; soft dorsal with irregular dark cross-bars. 

Wilhelm Land. 

Total length 93 to 166 mm. 

(5) Trematomus vicarius. 

Trematomus bernacchii subsp. vicarius, Lonnberg, Swedish South Polar Exped., Fish., p. 26 (1905). 
1 Notothenia duhia, Lonnberg, i.e., p. 28, pi. iii. fig. 9. 

Depth of body 3^ in the length, length of head 3^. Diameter of eye 4|- in length 
of head, interorbital width about 5. Maxillary extending to below anterior ^ of eye ; 





Fio. 6. — Head seen from above of A, Trematomus bernacchii, and B, T. mcarius. 

cheeks, opercles, and occiput scaly ; anterior part of interorbital region scaly in the 
middle; about 12 gill-rakers on lower part of anterior arch. Dorsal V, 33. Anal 31. 
Pectorals f , pelvics f the length of head. Caudal rounded. Caudal peduncle much 
deeper than long. 56 to 59 scales in a longitudinal series, 34 in upper lateral line ; 
lower lateral line without tubules. 

South Georo;ia. 

LoNNB erg's description is based on a single specimen, 240 mm. in length to base of 
caudal fin. 

Dr Lonnberg has kindly sent me a sketch of the upper surface of the head of the 
type, which is here reproduced, together with a figure of the head of a specimen of 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 2). 34 



262 MR C. TATE REGAN ON THE 

T. bernacchii of the same size. After comparing the type with examples of T. bernacchii 
that I sent him, Dr Lonnberg writes that the caudal peduncle is notably deeper and 
that the scales are larger. 

Dr Lonnberg has also sent me one of the types of N. dubia for examination. The 
specimen is 50 mm. in total length, and is a Trematomus, with the head scaled as in 
T. bernacchii, but with a smaller eye {3f in the length of head) and broader inter- 
orbital region (6 J in the length of head) than young examples of that species. I count 
Dorsal V, 37. Anal 32. 55 scales in a longitudinal series, 30 in the lateral line. 13 
gill-rakers on the lower part of anterior arch. It seems probable that this may be a 
young example of T. vicarius, although the interorbital region is scaleless. 

(6) Trematomus bernacchii. 

Bouleng., '^Southern Cross" Pisces, p. 181, pi. xiv. (1902). 

Notothenia elegans, Vail]., Exped. Antarct. Fran^aise, Poiss., p. 28 (1906). 

Depth of body 3 to 4| in the length, length of head 3-^ to 4. Diameter of eye 3 to 
4^ in the length of head, interorbital width 5 to 9. Maxillary extending to below 
anterior part or middle of eye ; occiput, cheeks, and opercles scaly ; interorbital region 
naked or with a median series of scales ; 13 to 15 gill-rakers on lower part of anterior 
arch. Dorsal IV-Vl, 34-38. Anal 31-35. Pectoral about f the length of head; 
pelvics just reaching anal in young, but not in adult. Caudal rounded. Caudal 
peduncle deeper than long. 60 to 75 scales in a longitudinal series from above base of 
pectoral fin to caudal, 30 to 42 in upper lateral line ; lower lateral line usually without 
tubules. Large dark spots in 2 or 3 alternating series ; upper ^ of spinous 
dorsal blackish. 

Graham Land and neighbouring islands ; Victoria Land. 

Here described from several specimens, 90 to 340 mm. in total length, including 
the types of the species from Cape Adare, 5 to 8 fathoms, and Duke of York Island, 
near Cape Adare, 3 to 4 fathoms [Southern Cross), examples from the Discovery winter 
quarters, Ross Island, and two from Station 325, Scotia Bay, South Orkneys, collected 
by the Scotia. 

In this species the interorbital region seems to be always scaleless in the young, and 
is often so in the adult fish. 

(7) Trematomus hansoni. 

Bouleng., ''Southern Cross" Pisces, p. 180, pi. xiii. (1902). 

Trematomus hansoni subsp. georgianus, Liiiinberg, Swedish South Polar Exped., Fisli., p. 25, pi. v. 

fig. 17 (1905). 
Notothenia siina, Vaill., Expe.il. Antarct. Franpaise, Poiss., p. 23 (1906). 

„ lepidorhinus (part.), Pappenheim, Deutsche Siidpolar- Exped., xiii., ZooL, v. p. 169 

(1912). 

Depth of body 3 J to 4f in the length, length of head 3| to 4. Diameter of eye 
3| to 5 in the length of head, interorbital width 5 to 6|. Maxillary extending to below 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 263 

anterior part or middle of eye ; occiput, interorbital region, cheeks, and opercles scaly ; 
13 to 16 gill-rakers on lower part of anterior arch. Dorsal V-VII, 36-41. Anal 
33-36. Pectoral f to |^ the length of head, longer than pelvics, which do not reach 
the vent. Caudal subtruncate. Caudal peduncle about as long as deep. 60 to 68 
scales in a longitudinal series from above base of pectoral fin to caudal, 38 to 46 in 
upper lateral line ; lower lateral line usually without tubules. Brownish, with large 
dark spots or cross-bars ; head often spotted ; fins usually barred with series of 
dark spots. 

South Georgia ; Graham Land ; Coats Land ; Wilhelm Land ; Victoria Land. 

Here described from several specimens, 160 to 380 mm. in total length, including 
the types of the species, from Cape Adare, 4 to 8 fathoms, and Duke of York Island, 
near Cape Adare, 3 to 4 fathoms [Southern Cross), examples from the Discovery 
winter quarters, Ross Island, and three from Coats Land, Station 411, 74° 01' S., 
22° 00' W., 161 fathoms ; temperature 28-9° F. ; trap ; March 1904. 

Dr Lonnberg's supposed subspecies from South Georgia is fully identical with the 
typical form. He gives the number of anal rays as (31) 32-33. but the figure shows 
36, and in an example that he has kindly sent me I count 35. There is no difference in 
the shape of the pectoral. 

Dr Pappenheim has kindly sent me for examination the smallest specimen of his 
N. lepidorhinus, 160 mm. in total length, which he has noticed as differing from the 
types in the larger number of dorsal rays (38 instead of 32 or 33). It differs also in the 
naked snout and preeorbital, shorter pelvic fins, lower lateral line without tubules, 
foramen enclosed in the hypercoracoid, etc., and is in every way similar to one of the 
types of T. hansoni, with which I have compared it. 

(8) Trematomus loennhergii, sp. n. (PL VIII. fig. 4.) 

Depth of body 4f to 5 in length, length of head 3 to 3|. Diameter of eye 3 to 4|- 
in the length of head, interorbital width 7 to 10. Maxillary extending to below 
anterior \ of eye ; upper surface of head to nostrils, cheeks, and opercles scaly ; 1 3 gill- 
rakers on lower part of anterior arch. Dorsal V-VI, 33-34. Anal 31-33. Pectoral 
as long as or a little shorter than head ; pelvics extending to origin of anal or beyond. 
Caudal rounded. Caudal peduncle longer than deep. 60 to 70 scales in a longitudinal 
series from above pectoral fin to caudal ; 36 to 42 in upper lateral line, which ends below 
posterior rays of dorsal ; lower lateral line without tubules. 

Victoria Land, Graham Land, and neighbouring islands. 

Here described from three specimens, two from the Discovery collection — the larger, 
132 mm. in total length, from south-west of the Balleny Islands, 254 fathoms; the 
smaller, 65 mm., in total length, from Tent Island, near Ross Island. The third 
example, also about 65 mm., is from Seymour Island, and has been sent to me for 
examination by Dr Lonnberg, who has recorded this species as Notothenia nicolai 



264 MR C. TATE REGAN ON THE 

{Swedish South Polar Exped., Fish., p. 45). I have named the species after Dr 
LoNNBERG in recognition of his kindness in sending me this and other specimens. 

2. PIeu7'agramma, Bouleng., 1902. 

"Southern Cross" Pisces, p. 187 (1902). 

Closely related to Trematomvs, differing especially in the very thin cycloid scales, 
the absence of pitted or tubular lateral line scales, and the feebly ossified skeleton, with 
parapopliyses developed only on the posterior praecaudal vertebrae. 

Coasts of the Antarctic Continent. 

BouLENGER has placed this genus in the family Leptoscopidse, but it has no affinity 
with Leptoscopus, and, on the other hand, is very near to Trematomus. A comparison 
of Pleuragramma antarcticum with Trematomus newnesii shows a very close agree- 
ment in external and internal characters, even to the number of fin-rays and vertebrae ; 
the pectoral arch is precisely similar. In Pleuragram,ma the two lateral lines are 
marked by scales with notched posterior edges, or, if the scales have been lost, by series 
of pores. 

Pleuragramm^a antarcticum. 

Bouleng., I.e., pi. xviii. ; Vaillant, Exped. Anturcf. Frangaise, Poiss., p. 48 (1906); Pappenheim, 
Deutsche Sildpolar- Exped., xiii., Zool., v. p. 164. 

Depth of body 5 to 6 in the length, length of head 3| to 4. Diameter of eye 3| to 
3f in the length of head, interorbital width 5 to 6. Lower jaw projecting ; maxillary 
extending to below anterior \ of eye ; upper surface of head naked ; cheeks and opercles 
scaly ; 20 to 25 gill-rakers on lower part of anterior arch. Dorsal VI-VII, 34-37. 
Anal 36-38. Pectoral truncated, i to f length of head. Caudal slightly emarginate. 
About 55 scales in a lateral longitudinal series. Silvery, back darker; sides and back 
powdered with blackish dots. 

Graham Land ; Wilhelm Land ; Victoria Land. 

Here described from several specimens, 150 to 200 mm. in total length, including 
the types of the species from Victoria Land [Southern Cross) and examples from near 
Cape Armitage, Ross Island, and from south-west of the Balleny Islands (Discovery). 
Vaillant gives D V, 39 ; A 38 ; Sq. 44, for specimens from Graham Land ; and Pappen- 
heim, D V-VIII, 34-38 ; A 36-38 ; Sq. 56-60, for examples from Wilhelm Land. 

3. Notothenia, Richards., 1844. 

"Erebus" and. " Terror" Fish., p. 5 ; Giinth., Cat. Fish., ii. p. 260 (1860). 
Macronotothen, Gill, Proc. Acad. Philad., 1861, p. 521. 

Differs from Trematomus only in that the hypercoracoid foramen is margined below 
by the hypocoracoid (fig. 3, 3, p. 249). 

Coasts of the Antarctic Continent and northwards to Patagonia, the Falkland 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 265 

Islands, South Sandwich Islands, Marion Islands, Kerguelen, southern New Zealand, 
and Chatham Islands. 

Although the interorbital region is relatively broader in the larger fish, its width, 
compared with the length of the head, is an important specific character. By the 
" interorbital width " is here understood the actual width of the osseous interorbital 
space ; this is nearly always less, and when narrow proportionately much less, than the 
interorbital width as measured by Boulenger. 

Synopsis of the Species. 
I. Opercles fully scaled. 

A. Upper surface and sides of head scaly, except snout and prseorbital ; inter- 

orbital width \ the length of head, or more. 

1. Upper lateral line of 62 to 65 tubular scales ; D VI, 32-34. A. 30-32. 

3 lateral lines . . . . . . . . .1. trigramma. 

2 lateral lines .......... 2. canina. 

2. Upper lateral line of 40 to 55 tubular scales. D V-VII, 32-37. A 31-35. 

a. 21 to 25 gill-rakers on lower part of anterior arch. D VII (VIII), 

34-36. A. 32-34 3. ramsayi. 

h. 14 to 19 gill-rakers on lower part of anterior arch. 

a. Upper lateral line not or scarcely extending beyond end of dorsal 

fin ; diameter of eye 4^ to 6 in length of head (in specimens up 

to 250 mm.). D VI-VII, 32-34. A. 31-34 4. tesselata. 

/3. Upper lateral line nearly reaching caudal fin. 

D VI-VII, 33-36. A 32-34. 16 to 19 gill-rakers on lower part of anterior 

arch. Eye 31 to 4f in length of head (in specimens of 125 to 250 mm.) 

5, iviltoni. 
D V (VI), 35-37. A 32-35. 16 to 19 gill-rakers on lower part of anterior 
arch. Eye 4 to 4-^- in length of head (in specimens of 90 to 180 mm.) 

6. brevicauda. 

D V-VI, 34-35. A 32-34. 14 or 15 gill-rakers on lower part of anterior arch. 

Eye 3 to 3| in length of head (in specimens of 130 to 180 mm.) 7. longijoes. 

3. Upper lateral line of 30 to 35 tubular scales; D V-VI, 28-31. 

A 27-30 8. sima. 

B. Upper surface and sides of head scaly, including prseorbital (and snout also 

except in N. scotti). 

1. Interorbital width 6 in length of head ; D VI-VII, 32-33. A 35-36. 

9. lepidorhinus. 

2. Interorbital width 9 to 1 3 in length of head. 

D IV-V, 36-37. A 32 10. squamifrons. 

D VI, 37-39. A 38 11. larseni. 

D V, 33. A 31 12. scotti. 



26G MR C. TATE REGAN ON THE 

C. Upper surface and sides of head scaly (except in N. nudifrons) ; prseorbital 
naked ; interorbital width y^Q the length of head, or less. 

1. Lower lateral line of 32 to 41 tubular scales; D VII-VIII, 31-33. 
A 31-33 18. gibberifrons. 

2. Lower lateral line of 15 to 18 tubular scales ; D VI-VII, 28-30. A 28-31. 

14. acuta. 

3. Lower lateral line without tubular scales. 

Upper surface of head scaly. D VII, 32. A 32 . . .15. vaillanti. 

Upper surface of head scaly. D IV- V, 35-37. A 33-35 . . 16. vnizops. 

Upper surface of head naked. D IV-VI, 37-39. A 34-36 . 17. nudifrons. 

II. Opercles scaly above, naked below ; upper surface of head scaly. 
Interorbital width 10 in length of head. D VII, 29. A 27 18. marionensis. 
Interorbital width 20 in length of head. D V-VI, 29-30. A 30-31. 

19. angustifrons. 

III. Opercles scaled only on upper part of operculum ; upper surface of head naked. 

A. Anal of 27-33 rays. 

a. Interorbital width 12 in length of head. D VI, 33. A 31. 

20. elegans. 
h. Interorbital width 4 to 7 in length of head. 
Cheek usually scaly behind eye. D IV-VI, 31-34. A 27-31 21. cornucola. 
Cheek scaly below and behind eye, its lower | (young) or \ (adult) naked. 

D IV-VI, 33-36. A 30-33 22. cyaneobrancha. 

Cheek scaly behind eye. D III-VII, 35-41. A 27-31 . 23. coiiiceps. 

c. Interorbital width 3| in length of head. D VI-VII, 33-35. A 27-29. 

24. rossii. 

B. Anal of 22 to 25 rays ; interorbital width 2| to 4| in length of head. 
D IV, 29-31. 48 to 56 scales in a lateral longitudinal series. 

25. macrocephala. 
D VI-VII, 28-29. 52 to 58 scales in a lateral longitudinal series. 

26. onicrolepidota. 
J) VI-VIIl, 26-27. 84 to 92 scales in a lateral longitudinal series 27. colbecki 

C. Anal of 18 to 20 rays ; interorbital width about 4 in length of head. 

28. Jilholi 

(1) Notothenia trigyximma, sp. n. (PL VI. fig. 2.) 

Depth of body 5 in the length, length of head 4. Diameter of eye 5 in the length 
of head and equal to the interorbital width. Lower jaw projecting ; maxillary 
extending to below anterior ^ of eye ; upper surface of head, except snout, cheeks and 
opercles scaly ; 15 gill-rakers on lower part of anterior arch. Dorsal VI, 34. Anal 32. 
Pectoral longer than pelvics, f as long as head, extending to above anal. Caudal 
rounded. About 85 scales in a lateral longitudinal series, 65 in upper lateral line, 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 267 

which nearly reaches caudal, 13 in line on middle of tail, and 40 to 45 in a third lower 
lateral line, which is separated by 4 or 5 longitudinal series of scales from the base of 
the anal fin. Brownish ; fins darker. 

Port Stanley, Falklands. 

Total length 280 mm. 

I was at first inclined to make this species the type of a new genus, but on examining 
related species of Nototlienia I found a specimen of N. hrevicauda with a third lateral 
line on one side only, formed of 10 tubular scales and separated from the posterior part 
of the anal fin by 3 series of scales. 



(2) Notothenia canina. 
Smitt, Bill. Sv. Vet.-Akad. HaiidL, xxiii., iv., No. 3, p. 32, pi. ii, fig. 2:2 (1897). 

Evidently closely related to N. tesselata, but the outer series of teeth stronger, 
spaced, more canine-like, and the upper lateral line with 62 to 65 tubular scales. 
Dorsal VI, 32-33. Anal 30-31. 

East coast of Patagonia. 

Total length 138 mm. 

Notothenia acuta, Steind. {non Giinther) (Zool. Jahrh., Suppl. iv., 1897-8, p. 303), 
from Chile, is probably closely related to N. canina. 



(3) Notothenia ramsayi, sp. n. (PL VII. fig. 1.) 

Depth of body 4 to 5-^- in the length of the fish, length of head about 3^. 
Diameter of eye 4 to A}^ in the length of head, interorbital width 4-^ to 7. Jaws 
equal anteriorly ; maxillary extending to below anterior ^ of eye ; cheeks, opercles, and 
upper surface of head, to between the nostrils, scaly; 21 to 25 gill-rakers on lower 
part of anterior arch. Dorsal VII (VIII), 34-36. Anal 32-34. Pectoral from less 
than I to f length of head ; pel vies as long, extending to vent or to anal fin. Caudal 
rounded or subtruncate. Caudal peduncle as long as deep, or deeper than long, its 
least depth ^ to f the length of head. 60 to 72 scales in a longitudinal series, from 
above base of pectoral to caudal fin ; 46 to 54 in upper lateral line, which almost 
reaches the caudal ; 8 to 1 7 in lower lateral line. A lateral series of 5 to 7 dark blotches 
or vertical bars. 

Several specimens, 200 to 300 mm. in total length, taken on 1st December 1903 
from the Burdwood Bank, Scotia Station 346, 54° 24' S., 50° 32' W. ; depth 56 
fathoms; surface temperature 41 '8° F. ; one from Isthmus Bay, Magellan Straits, 
14 fathoms (Coppinger). 

This species is named in memory of Allan George Ramsay, chief engineer of the 
Scotia, who died at Scotia Bay, South Orkneys, on 6th August 1903. 



268 MR C. TATE REGAN ON THE 

(4) Notoihenia tesselata. 

Richards., " Erelus" and "Terror" Fif-h., p. 19, pi. xii. figs. 3, 4 (1845); Giintli., Cat. Fish., 

ii. p. 260 (1860). 
Notothenia veitchii, Giinth., Anri. Mag. Nat. Hist. (4), xiv., 1874, p. 370. 

„ brevipes, Lonnberg, Swedish South Polar Exped., Fish., p. 15 (1905). 

Depth of body 4-^ to G in the length, length of head 3^ to 3f . Diameter of eye 
4^ to 6 in the length of head, interorbital width 5-^- to 6. Lower jaw rather prominent; 
maxillary extending to below anterior part or middle of eye ; cheeks, opercles, and 
upper surface of head, except snout, scaly; 14 to 16 gill-rakers on lower part of 
anterior arch. Dorsal VI-VII, 32-34. Anal 31-34. Pectoral from less than 
I to more than f the length of head, usually longer than pel vies, which seldom 
reach the anal. Caudal rounded. Caudal peduncle deeper than long. 62 to 
78 scales in a longitudinal series from above base of pectoral to caudal fin, 41 to 48 
in upper lateral line, which ends below or a little behind end of dorsal fin, 6 to 11 
in lower lateral line. Body marbled ; spinous dorsal dusky, pale at the base ; soft 
dorsal, caudal, and sometimes anal, with series of dark spots. 

Chile ; Magellan Straits ; Falkland Islands. 

Here described from several examples, 140 to 250 mm. in total length, from the 
Falkland Islands, Magellan Straits, and Chile, including the types of the species and 
specimens from Station 118, Port Stanley, Falkland Islands, collected by the Scotia. 
In young specimens {N. veitchii, N. hrevipes) the interorbital width is ] or ^r the length 
of head. 

(5) Notothenia wiltoni, sp. n. (PI. VII. fig. 2.) 

Depth of body 4| to 5-^ in the length of the fish, length of head S^ to 3|. 
Diameter of eye 3| to 4f in the length of head, interorbital width 6 to 7. Jaws equal 
anteriorly ; maxillary extending to below anterior ^ of eye or beyond ; cheeks, opercles, 
and upper surface of head, except snout, scaly; 16 to 19 gill rakers on lower part of 
anterior arch. Dorsal VI-VII, 33-36. Anal 32-34. Pectoral | or f the length of 
head ; pelvics as long or somewhat longer, extending to vent or to anal fin. Caudal 
rounded or subtruncate. Caudal peduncle deeper than long, its least depth | to ^j the 
length of head. 62 to 70 scales in a longitudinal series from above base of pectoral 
to caudal fin, 47 to 54 in upper lateral line, which almost reaches caudal, 7 to 14 in 
lower lateral line. Body with irregular dark cross-bars ; spinous dorsal dusky, pale 
at base. 

Ten specimens, 125 to 250 mm. in total length — one from Orange Bay (Paris Mus.) ; 
another from the Straits of Magellan (Coppinger) ; the others taken by the Scotia at 
Port Stanley (Station 118) and Port William (Station 349), Falkland Islands (shore,' 
51° 41' S., 57° 51' W,), and on the Burdwood Bank (Station 346, 54° 25' S., 57° 32' W., 
56 fathoms ; surface temperature 41"8° F. ; otter trawl; 1st December 1903). 

This species is named after Mr D. W. Wilton, zoologist of the Scotia. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 269 

(6) Notothenia hrevicauda. 

Ldnnberg, Sivedish South Polar Exped., Fish., p. 6, pi. v. fig. 16 (1905). 

Depth of body 4^ to 5 in the length of the fish, length of head 3^ to 4. Diameter 
of eye 4 to 4|- in the length of head, interorbital width 7 or 8. Maxillary extending 
to below anterior ^ of eye ; cheeks, 'opercles, and upper surface of head, except snout, 
scaly; 16 to 19 gill-rakers on lower part of anterior arch. Dorsal V (VI), 35-37. 
Anal 32-35. Pectoral f to f the length of head ; pelvics as long or a little longer, 
extending to the anal. Caudal rounded. Caudal peduncle \ to f as long as deep, its 
least depth about \ the length of head. 60 to 70 scales in a longitudinal series from 
above base of pectoral to caudal fin, 44 to 50 in upper lateral line, which ends 2 to 4 
scales in front of the caudal, 4 to 12 in lower lateral line. Body with irregular dark 
cross-bars ; pectorals yellow ; pelvics, anal, spinous dorsal, and base of soft dorsal dusky. 

Magellan Straits ; Falkland Islands. 

Twelve specimens, 90 to 180 mm. in total length, including examples from Port 
Stanley (June 1903, 9-10 fathoms) and from Port William, Falkland Islands (January 
1903, 6 fathoms). Lonnberg's type, a specimen of 120 mm., came from Tierra del 
Fuego, and there are examples from that locality in the British Museum. 

(7) Notothenia longipes. 

Steind., (S'ifeMra(/s&. Akad. Wien, Ixxii., 1876, p. 70, pi. vi. fig. 7; Giintb., ^^ Challenger" Shore 
Fish., p. 21 (1880). 

Depth of body 5^ to 6t} in the length, length of head 3^ to 3f . Diameter of eye 

3 to 3^ in the length of head, interorbital width 7 or 8. Maxillary extending to below 
anterior \ of eye ; upper surface and sides of head, except snout and prseorbital, scaly ; 
14 or 15 gill-rakers on lower part of anterior arch. Dorsal V-VI, 34-35. Anal 32-34. 
Pectoral | to f the length of head, somewhat shorter than pelvics, which reach the anal. 
Caudal rounded. Caudal peduncle f to | as long as deep, its least depth from ^ to |^ 
the length of head. 62 to 70 scales in a longitudinal series, 46 to 55 in upper lateral 
line, which almost reaches caudal, 6 to 1 3 in lower lateral line. Body with irregular 
brownish cross-bars. 

Patagonia and Magellan Straits. 

Here described from four examples, 130 to 180 mm. in total length. 

(8) Notothenia sima. 

Richards., ''Erebus" and " Terror" Fish., p. 19, pi. xi. fig. 1 (1845) ; Giinth., Cat. Fish., ii., p. 262 (1860). 
Notothenia squamiceps, Peters, Monatsb. Akad. Berlin, 1876, p. 837. 

„ karlandrem, Lbnnberg, Swedish Soidh Polar Exped., Fish., p. 14, pi. iv. fig. 13 (1905). 

Depth of body 4 to 5 in the length, length of head 3^ to 3|. Diameter of eye 

4 to 5 in the length of head, interorbital width 6 to 8. Maxillary extending to below 
anterior part or middle of eye ; occiput, interorbital region, cheeks, and opercles scaly ; 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 2). 35 



270 MR C. TATE REGAN ON THE 

10 to 12 gill-rakers on lower part of anterior arch. Dorsal V-VI, 28-31. Anal 27-30. 
Pectoral f to f the length of head, about as long as pelvics, which reach the vent. 
Caudal rounded. Caudal peduncle much deeper than long. 40 to 46 scales in a 
longitudinal series from above base of pectoral fin to caudal, 30 to 35 in upper lateral 
line, which ends below posterior rays of dorsal, 2 to 12 in lower lateral line, when 
developed. Body with irregular dark cross-bars ; vertical fins more or less dusky, the 
caudal often barred and with 2 or 3 dark spots at the base. 

Magellan Straits ; Falkland Islands. 

Here described from several specimens, 60 to 120 mm. in total length, including 
the type of the species, from the Falkland Islands, and a co-type of N. karlandrew. 

The Scotia examples are from Station 118, Port Stanley, Falklands, 51° 41' S., 
57° 51' W., and there are others in the British Museum collection from Magellan. 

(9) Notothenia lepidorhinus. 

Notothenia lepidorhinus (part.), Pappenheim, Deutsche Siidpolar-Exped., xiii., Zool., v. p. 169, 
pi. ix. fig. 1 and pi. x. fig. 1 (1912). 

Depth of body 4 to A^ in the length, length of head 3^ to 3f. Diameter of eye 
3 to 3-^- in the length of head, interorbital width about 6. Maxillary extending to 
below anterior margin of pupil ; upper surface and sides of head, including snout and 
prseorbital, scaly ; 16 gill-rakers on lower part of anterior arch. Dorsal VI-VII, 32-33. 
Anal 35-36. Pectoral |- the length of head ; pelvics extending beyond origin of anal. 
Caudal rounded. Caudal peduncle somewhat deeper than long. 72 to 82 scales in a 
longitudinal series, 45 to 56 in upper lateral line, 32 to 38 in lower lateral line. Body 
with irregular dark cross-bars ; spinous dorsal dark anteriorly ; soft dorsal with dark 
oblique stripes. 

Wilhelm Land, 385 metres. 

The types measure 186 to 240 mm. in total length. 

(10) Notothenia squamifrons. 

Giinth., " Challenger" Shore Fish., p. 16, pi. viii. fig. C (1880). 

Depth of body 4^ in the length, length of head 3|. Diameter of eye 3 to 3^ in the 
length of head, interorbital width 9 to 12. Maxillary extending to below anterior \ of 
eye ; upper surface and sides of head, including snout and prseorbital, scaly ; 14 to 16 
gill-rakers on lower part of anterior arch. Dorsal IV-V, 36-37. Anal 32. Pectoral 
f the length of head, rather shorter than pelvics, which reach the anal. Caudal 
peduncle deeper than long. 55 scales in a longitudinal series from above base of 
pectoral to caudal, 44 or 45 in upper lateral line, which ends below end of dorsal, or 
just behind it, 15 to 18 in lower lateral line. Body with broad irregular cross-bars; 
cheek with two oblique stripes ; spinous dorsal blackish. 

Kerguelen. 

Here described from the types, two specimens, 110 and 150 mm. in total length. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 271 

(11) Notothenia larseni. 
Lonnberg, Sivedish South Polar Exped., Fish., p. 31, pi. i. fig. 3 (1905). 

Depth of body 4^ to 5 in the length, length of head 3|. Diameter of eye 3 in 
length of head, interorbital width 11 to 13. Maxillary extending a little beyond 
vertical from anterior margin of eye ; upper surface and sides of head entirely scaly. 
Dorsal VI, 37-39. Anal 38. Pectoral a little shorter than head, longer than pelvics, 
which just reach anal. Caudal rounded. Caudal peduncle as long as deep. 69 to 7Q 
scales in a longitudinal series above upper lateral line, which has 55 or 56 tubes and 
nearly reaches caudal ; lower lateral line without tubes. Body with irregular cross-bars ; 
dorsal with oblique series of spots. 

South Georgia; length 178 mm. 

(12) Notothenia scotti. 
Bouleng., Nat. Antard. Exped. Nat. Hid., ii. Fi?h., p. 2, pi. i. fig. 1 (1907). 

Depth of body b^ in the length, length of head 3^. Diameter of eye 2| in the 
length of head, interorbital width 12. Maxillary extending to below anterior ^ of eye ; 
upper surface of head, except snout, and sides of head, including praeorbital, scaly ; 
12 gill-rakers on lower part of anterior arch. Dorsal V, 33. Anal 31. Pectoral f the 
length of head, somewhat shorter than pelvics, which reach anal. Caudal peduncle as 
long as deep. 50 scales in a longitudinal series from above base of pectoral to caudal, 
probably about 40 in upper lateral line. Body with irregular cross-bars ; spinous 
dorsal blackish ; soft dorsal and anal blackish posteriorly. 

Near Edward Land. 

Here described from the type, a specimen of 110 mm. taken at a depth of 300 
fathoms off the Ross Barrier, 27th January 1902. In the original description and figure 
the fin-rays are miscounted. 

(13) Notothenia gihberifrons. 

Lonnberg, Sivedish South Polar Exped., Fish., p. 33, pi. iii. fig. 10 (1905); Yaillant, Exped. 
Antard. Fran^aise, Poiss., p. 33 (1906). 

Depth of body 5 to 5 J in the length, length of head 3^ to 3f. Diameter of eye 
4 to 4f in the length of head, interorbital width 12 to 16. Jaws equal anteriorly; 
maxillary not or barely reaching vertical from anterior margin of eye ; cheeks, opercles, 
and upper surface of head to nostrils scaly ; 1 gill-rakers on lower part of anterior arch. 
Dorsal VII-VIII, 31-33. Anal 31-33. Pectoral | the length of head ; pelvics f to f 
length of head, not reaching vent. Caudal truncate. Caudal peduncle nearly as long 
as deep ; 55 to 66 scales in a longitudinal series from above base of pectoral to caudal, 36 
to 44 (to bXjide Lonnberg) in upper lateral line, which ends below posterior part of dorsal, 
32 to 41 in lower lateral line. Upper part of body irregularly spotted ; dorsal, caudal, 
and pectoral fins with series bf dark spots. A water-colour drawing shows the ground 
colour yellow, the fins greenish, the spots brown. 



272 MR C. TATE REGAN ON THE 

Graham Land ; South Georgia ; South Orkneys ; South Shetlands. 

Here described from six specimens, 280 to 340 mm. in total length, taken in July 
1903 at Station 325, Scotia Bay, 27 fathoms, and Station 326, Jessie Bay, 10 fathoms, 
South Orkneys ; there are also two quite small specimens from the same locality. 
Lonnbekg's types come from South Georgia, and there is a specimen in the British 
Museum from the South Shetlands. 



(14) Notothenia acuta. (PI. VIII. fig. 3.) 

Giinth., " GJiallenger " Shore Fish., p. 17 (1880); Pappenheim, Deutsche Sildpolar-Exped., \m., 
Zool, V. p. 171, pi. ix. fig. 3 (1912). 

Depth of body 6 in the length, length of head 3^. Diameter of eye 3f in the 
length of head, interorbital width 16. Maxillary extending to below anterior ^ of eye ; 
sides and upper surface of head scaly, except snout and prseorbital ; 12 gill-rakers on 
lower part of anterior arch. Dorsal VI (Vll), (28-29) 30. Anal (28-30) 31. Pectoral 
nearly as long as the head, longer than pelvics, which reach the vent. Caudal peduncle 
somewhat deeper than long. 60 scales in a longitudinal series from above base of 
pectoral to caudal, 38 in upper lateral line, which ends below posterior part of dorsal, 
16 to 18 in lower lateral line. Body marbled ; dorsal rays with series of small spots; 
caudal barred. 

Kerguelen, 

Here described from the type, about 62 mm. in total length, from Kerguelen. 



(15) Notothenia vaiUanti, n. sp. 
Notothenia acutd (nan Giinth.), Vaillanfc, Exped. Antard. Franr/xise, Poise., p. 31 (1906). 

Depth of body 5| in the length, length of head 3^. Diameter of eye 3 in the length 
of head, interorbital width 14. Maxillary extending to below anterior ^ of eye; 
sides and upper surface of head scaly, except snout and prseorbital; 10 gill-rakers on 
lower part of anterior arch. Dorsal VII, 32. Anal 32. Pectoral a little shorter 
than head, as long as pelvics, which reach the anal. Caudal peduncle a little longer 
than deep. 55 scales in a longitudmal series from above pectoral fin to caudal, 34 in 
upper lateral line, which ends below posterior part of dorsal ; lower lateral line without 
tubular scales. Body with irregular cross-bars, broken up into 3 or 4 series of 
alternating spots ; dorsal with small spots ; caudal barred. 

Graham Land ; Booth, Wandel, and Wiencke Islands. 

Here described from a specimen of 56 (46 + 10) ram. Measurements of this example 
are given by Vaillant {t.c, p. 32), and also those of a much larger fish, 410 mm. in 
length to base of caudal, with the eye I and the interorbital width y\j^ of the length 
of the head. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 273 

(16) Notothenia mizops. 

Gunth., " Challenger" Shore Fish., p. 16, pi. viii. fig. D (1880). 

Depth of body 4-^ to 4f in the length, length of head 3f to 4. Diameter of eye 
3 to 3| in the length of head, interorbital width about 15. Maxillary extending to 
below anterior ^ or ^ of eye ; cheeks, opercles, occiput, and interorbital region scaly ; 
9 to 13 gill-rakers on lower part of anterior arch. Dorsal lY-V, 35-37. Anal 33-35. 
Pectoral f to f the length of head ; pelvics longer, reaching the anal. Caudal rounded. 
Caudal peduncle deeper than long. 48 to 55 scales in a longitudinal series from above 
base of pectoral to caudal, 33 to 38 in upper lateral line, which ends below posterior 
part of dorsal ; lower lateral line without tubular scales. Body with 2 series of 
large, partly confluent, irregular blackish spots ; cheek with 2 oblique stripes ; a 
blackish spot on spinous dorsal ; vertical fins with or without series of dark spots. 

Kerguelen. 

Here described from the types, five specimens, 70 to 170 mm. in total length. 



(17) Notothenia nudifrons. 

Notothenia mizops var. nudifrons, Loniiberg, Swedish South Polar Exped., Fish., p. 30, pi. i. 

fig. 2 (1905). 
Notothenia mizops, Vaillant, Exped. Antard. Frangaise, Poiss., p. 30 (1906). 

Closely related to N. mizops, but occiput and interorbital region naked, fin- 
rays usually more numerous (Dorsal IV-VI, 37-39. Anal 34-36), and scales smaller, 
55 to 65 in a longitudinal series ; 11 or 12 gill-rakers on lower part of anterior arch. 
Coloration of N. mizops. A water-colour drawing shows the fish reddish, the spots 
dark brown. 

South Georgia ; South Orkneys ; Graham Land. 

Here described from nine specimens, 70 to 150 mm. in total length, from Station 
325, Scotia Bay, South Orkneys, depth 9 to 10 fathoms (June 1903), from South 
Georgia (Swedish Expedition), and from Graham Land (Paris Mus.). 

18. Notothenia marionensis. (PI. VIIL, fig. 2.) 

Giinth., " Challenger" Shore Fish., p. 17 (1880). 

Depth of body 5 in the length, length of head 3f . Diameter of eye 4 in the length 
of head, interorbital width 10. Jaws equal anteriorly; maxillary extending to below 
anterior ^ of eye ; scales on upper half of cheek and opercles, on interorbital region and 
occiput ; a transverse naked strip separating last from scales of nape ; 11 gill-rakers on 
lower part of anterior arch. Dorsal VII, 29. Anal 27. Pectoral fin f , pelvic f the 
length of head. Caudal rounded or subtruncate. Caudal peduncle deeper than long. 
48 scales in a longitudinal series from above base of pectoral to caudal fin ; 35 in upper 
lateral line, which ends below posterior part of dorsal ; 16 in lower lateral line. Body 



274 MR C. TATE EEGAN ON THE 

with irregular dark spots ; a blackish spot on upper part of base of pectoral ; dorsal 
and caudal with series of small spots. 

Marion Island. 

Here described from the type, 82 mm. in total length, from Marion Island, 
50 to 75 fathoms. 

(19) Notothenia migustifrons. (PI. VIII. fig. 1.) 
Fischer, Jahrb. Hamburg Wiss. Anst., ii., 188.5, p. 55. 

Depth of body 5 in the length, length of head 3f . Diameter of eye 4 in the length 
of head, interorbital width about 20. Maxillary extending to below anterior margin 
or anterior ^ of eye ; upper surface of head scaly to between nostrils ; cheeks and 
opercles in great part scaly, but naked below ; 10 or 11 gill-rakers on lower part of 
anterior arch. Dorsal V-VI, 29-30. Anal 30-31. Pectoral nearly as long as head, 
longer than pelvics. Caudal rounded. Caudal peduncle about as long as deep. 
46 to 52 scales in a longitudinal series from above base of pectoral to caudal, 26 to 33 
in upper lateral line, which ends below middle or posterior part of soft dorsal, 16 to 
23 in lower lateral line. Dark bars across the back, which break up into spots on the 
sides of the body ; often a bar through spinous dorsal connecting the bases of the 
pectorals ; dorsal, caudal, and pectoral fins with series of small dark spots on the rays ; 
pelvics and anal pale, sometimes with a few spots. 

South Georgia ; South Sandwich Islands. 

Here described from six specimens, 70 to 116 mm. in total length, one from 
South Georgia (Lonnberg), the rest from the South Sandwich group (Allardyce). 

(20) Notothenia elegans. 

Giinth., " Challenger" Shore Fish., p. 21, pi. xi. fig. C (1880). 

Depth of body 6 to 7 in the length, length of head 4^. Diameter of eye 3| in 
the length of head, interorbital width 12. Maxillary extending to below anterior 
^ of eye ; a few scales behind eye and on upper part of operculum, rest of head 
probably scaleless ; 10 gill-rakers on lower part of anterior arch. Dorsal VI, 33. 
Anal 31. Pectoral f the length of head, rather shorter than the pelvics, which reach 
the anal. Caudal rounded. Caudal peduncle somewhat deeper than long. 46 to 48 
scales in a longitudinal series from above base of pectoral to caudal, 40 or 41 in upper 
lateral line, which ends below last rays of dorsal, 4 to 9 in lower lateral line. Large 
dark spots or vertical bars on sides of body ; tip of spinous dorsal pink ; soft dorsal 
with 3 or 4 series of small dark spots. 

Magellan Straits. 

Here described from the types, two specimens 95 mm. in total length, from off 
Cape Virgins, Patagonia, 55 fathoms. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 275 



(21) Notothenia cornucola. 

Richards., "Urebiis" and "Terror" Fish., pp. 8, 18, pis. viii. figs. 4, 5, and xi. figs. 3, 4 (1845) ; 

Giinth., Cat. Fish., ii. p. 261 (1860). 
Notothenia vinjata, Richards., t.c, p. 18, pi. xi. figs. 5, 6 ; Giinth., t.c, p. 262. 

,, maryinata, Richards., t.c, p. 18, pi. xii. figs. 3, 4. 

„ modesta, Steind., Zool. Jahrb. Suppl., iv., 1898, p. 302, pi. xx. fig. 3. 

Depth of body 3f to 4| in the length, length of head 3 to 3J. Diameter of eye 
4| to 5 in the length of head, interorbital width 5| to 7. Jaws equal anteriorly ; 
maxillary extending to below middle of eye ; usually a few scales behind eye and on 
upper part of operculum ; 11 or 12 gill-rakers on the lower part of anterior arch. 
Dorsal IV-VI, 31-34. Anal 27-31. Pectoral about f the length of head, extending 
to above origin of anal or a little beyond ; pelvics about as long. Caudal rounded. 
Caudal peduncle much deeper than long. 47 to 55 scales in a longitudinal series from 
above base of pectoral to caudal, 36 to 42 in upper lateral line, which ends below last 
2 or 3 rays of dorsal ; 6 to 12 in lower lateral line, when it is developed ; only 2 or 3 
scales between lateral line and posterior rays of dorsal. Body usually spotted or 
marbled, sometimes with a pale lateral band ; vertical fins dusky. 

Patagonia ; Magellan Straits ; Falkland Islands ; New Zealand ; Chatham Islands. 

Here described from numerous specimens, 90 to 140 mm. in total length, including 
the types of the species, of N. virgata and of N. marginata, mostly from the Falkland 
Islands and Magellan Straits ; one small specimen from New Zealand. Some examples 
were taken at Station 118, Port Stanley, Falkland Islands, shore, by the Scotia. 



(22) Notothenia cynaneohrancha. 

Richards., "Erebus" and "Terror" Fish., p. 7, pi. iv. (1844); Gunth., Cat. Fish., ii. p. 261 (1860). 
Notothenia purpuriceps, Richards., I.e., pi. ii. figs. 3, 4; Giinth., t.c, p. 262. 

Depth of body 4 to 5 in the length, length of head 3 to 4. Diameter of eye 
4 to 6 in the length of head, interorbital width 5 to 6. Jaws equal anteriorly ; 
maxillary extending to below middle or posterior part of eye ; upper surface of head 
naked except for a few temporal and post-temporal scales, which may be absent in the 
young ; cheek scaly behind and below eye, the lower | (young) or ^ (adult) naked ; 
upper part of operculum scaly ; 10 to 13 gill-rakers on lower part of anterior arch. 
Dorsal IV-VI, 33-36. Anal 30-33. Pectoral | the length of head, extending to 
above vent or origin of anal (adult) or a little beyond (young) ; pelvics about as long. 
Caudal rounded. Caudal peduncle much deeper than long. 60 to 70 scales in a 
longitudinal series from above base of pectoral to caudal, 32 to 39 in upper lateral 
line, which ends below posterior part of dorsal ; lower lateral line, when developed, 
with 6 to 15 tubular scales. A dark oblique stripe from eye to angle of prseoperculum, 
another below it. 



276 MR C. TATE REGAN ON THE 



Kerojuelen. 



Here described from several specimens, 120 to 260 mm. in total length, including 
the typo of the species. 

(23) Notothenia coi'iiceps. 

Richards., "Erebus" and "Terror" Fish., p. 5, pi. iii. figs. 1, 2 (1884); Gunth., Cat. Fish., ii. 
p. 261 (1860) ; Vaill., Expkl. Antard. Fran^aise, Poiss., p. 24 (1906). 

Depth of body 3f to i^ in the length, length of head 3 to 3|. Diameter of eye 
4^ to 7 in the length of head, interorbital width 4 to 5. Jaws equal anteriorly ; 
maxillary extending to below middle (young) or posterior margin (adult) of eye ; head 
naked except for a few scales behind eye, on upper part of operculum, and on post- 
temporal region; 10 to 14 gill-rakers on lower part of anterior arch. Dorsal III-VII, 
35-40. Anal 27-31. Pectoral from less than f (in large specimens) to f the length 
of head, extending to above origin or anterior rays of anal ; pelvics shorter, not or 
barely reaching vent. Caudal subtruncate. Caudal peduncle nearly as long as deep. 
54 to 68 scales in a longitudinal series from above base of pectoral to caudal, 34 to 49 
in upper lateral line, which ends below posterior part of dorsal, 8 to 17 in lower lateral 
line ; 4 or 5 scales between lateral line and posterior dorsal rays. Colour varying from 
dark greenish black to a pale orange, with or w^ithout spots or marking ; usually one 
or two oblique dark bars across cheek, sometimes broken up into spots ; head sometimes 
with pale spots enclosed in dark rings ; spots on body and dorsal fin sometimes large 
and tesselated, more often smaller and scattered, rarely uniting to form longitudinal 
stripes ; soft dorsal and anal usually with a pale edge. 

Graham Land and neighbouring islands ; Kerguelen ; Victoria Land. 

Here described from a large series of specimens obtained by the Scotia at Station 325, 
Scotia Bay, South Orkneys, 160 to 450 mm. in total length, in addition to the type 
of the species from Kerguelen, examples from Duke of York Island and Cape Adare, 
Victoria Land {Southern Cross), and from Graham Land (Frangais). 

(24) Notothenia rossii. 

Richards., "Erebus" and "Terror" Fish., p. 9, pi. v. figs. 1, 2 (1844); Giinth., Cat. Fish., ii. 

p. 263 (1860). 
Macronotothen rossii, Gill, Proe. Acad. Philad., 1861, p. 521. 
Notothenia marmorata, Fischer, Jahrh. Havib. Wiss. Anst., ii., 1885, p. 53; Lonnberg, Swedish 

South Polar Exped., Fish., p. 34. 

Depth of body 4 j- to 4| in the length, length of head 3| to 3f. Diameter of eye 
5 to 5^ in the length of head, interorbital width 3^. Jaws equal anteriorly ; maxillary 
extending to below anterior margin of pupil ; scales on upper part of cheek and operculum 
and on temporal region ; upper surface of head papillose ; 12 gill-rakers on lower part 
of anterior arch. Dorsal V-VH, 33-35. Anal 27-29. Pectoral | the length of head, 
longer than pelvics. Caudal truncate. Caudal peduncle as long as or a little longer 
than deep. 58 to 62 scales in a longitudinal series from above base of pectoral fin to 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 277 

caudal, 46 to 52 in upper lateral line, which ends below posterior rays of dorsal, 10 
to 18 in lower lateral line. Body marbled; dorsal with 2 or 3 series of dark spots ; 
anal and caudal with a dark band. 

South Georgia ; South Orkneys. 

Here described from two specimens, 250 mm. in total length, from South Georgia ; 
a little fish, 62 mm. in total length, obtained by the Scotia at Station 325, Scotia 
Bay, South Orkneys, seems to belong to this species. 

Notothenia rossii was based on a large stuffed specimen, 850 mm. in total length, 
with the dorsal spines short and blunt as they often are in large Nototheniids. 

(25) Notothenia macrocephala. 

Giiuth., Cat. Fish., ii. p. 263 (1860). 

Notothenia maoriensis, Haast, Trans. N.Z. Inst., v., 1873, p. 276, pi. xvi. fig. 

,, angudata, Hutton, Ann. Mag. Nat. Hist. (4), xvi., 1875, p. 315. 

„ hassleriana, Steind., Sitzungsb. Akad. Wien, Ixxii., 1876, p. 69, pi. vi. fig. 

., antarctica, Peters, Monatsb. Akad. Berlin, 1876, p. 837. 

,, arguta, Hutton, Trans. N.Z. Inst., xi., 1879, p. 339. 

,, porteri, Delfin, Bev. Ghilen. Hist. Nat., iii., 1899, p. 117. 

Depth of body 3 to 4 in the length, length of head 3^ to 3f. Diameter of eye 
4 to 6 in the length of head, interorbital width 2-| to 3|. Jaws equal anteriorly ; 
maxillary extending to below anterior \ of eye ; imbricate scales behind eye and on 
upper part of operculum ; upper surface and sides of head otherwise naked, papillose ; 
10 to 12 gill-rakers on lower part of anterior arch. Dorsal IV (III- VI), 29-31. Anal 
22-25, Pectoral | to f the length of head, considerably longer than pelvics. Caudal 
truncate or slightly emarginate. Caudal peduncle usually somewhat longer than deep, 
48 to 56 scales in a longitudinal series from above base of pectoral fin to caudal; 36 
to 44 in upper lateral line, which ends below posterior rays of dorsal, 6 to 12 in lower 
lateral line. More or less distinct longitudinal stripes or series of spots on the sides ; 
dorsal dusky, sometimes reticulated ; caudal, anal, and pelvics sometimes similarly 
coloured. 

Patagonia ; Magellan Straits ; Falkland Islands ; New Zealand ; Auckland Island ; 
Campbell Island. 

Here described from several specimens, 130 to 280 mm. in total length, from New 
Zealand, Campbell Island, Magellan Straits, and the Falkland Islands. In addition 
to the type of the species, types of N. arguta and N. angustata have been examined. 

(26) Notothenia microlepidota. 

Hutton, Trans. N.Z. Inst., viii., 1876, p. 213; Waite, Subantardic Isl. N. Zealand, Pisces, 

p. 590, fig. (1909). 
Notothenia parva, Hutton, Trans. N.Z. Inst., xi., 1879, p. 339. 

Depth of body 4 to 5 in the length, length of head 3^ to 3-^. Diameter of eye 
4 J to 6^ in the length of head, interorbital width 3^ to 4^-. Upper surface of head 
TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 2). 36 



278 MR C. TATE REGAN ON THE 

naked, papillose ; sides mostly naked, scaly behind the eye and on upper part of 
operculum ; 11 or 12 gill-rakers on lower part of anterior arch. Dorsal VI-VII, 28-29. 
Anal 23-25. Pectoral ^- (adult) to f (young) the length of head ; pelvics nearly as 
long, not reaching vent. Caudal rounded or subtruncate. Caudal peduncle deeper 
than long. 52 to 58 scales in a longitudinal series from above base of pectoral fin to 
caudal, 52 to 60 in upper lateral line, which ends near end of dorsal fin, 10 to 15 in 
lower lateral line. Head reticulated ; body and fins spotted. 

New Zealand ; Auckland Island ; Campbell Island. 

Here described from five specimens, 90 to 500 mm. in total length, from Auckland 
and Campbell Islands, the smallest the type of N. parva. 

(27) Notothenia colhecki. 

Bouleng., "Southern Cross" Pisces, p. 185, pi. xvi. (1902) ; Waite, Suhantarctic Isl. N. Zealand, 
Pisces, p. 594 (1909). 

Depth of body 4 to 5 in the length, length of head 3^ to 3f. Diameter of eye 
5 to 7 in the length of head, interorbital width 3 to 4. Maxillary extending to below 
anterior part or middle of eye ; head mostly naked, with granular papillae, scaly behind 
the eye and on upper part of operculum ; 15 to 18 gill-rakers on lower part of anterior 
arch. Dorsal VI-VIII, 26-27. Anal 23-24. Pectoral | to | the length of head, some- 
what longer than pelvics, which do not reach the vent. Caudal emarginate. Caudal 
peduncle longer than deep, 84 to 92 scales in a longitudinal series from above base of 
pectoral to caudal, 62 to 69 in upper lateral line, which ends near end of dorsal fin, 
24 to 35 in lower lateral line. Brownish above, yellowish below ; a pair of oblique 
stripes across the cheek ; dorsal and caudal dusky. 

Auckland Island ; Campbell Island. 

Here described from specimens 125 to 550 mm. in total length, including the types 
from Campbell Island and a large stuffed specimen from Auckland, 

(28) Notothenia Jilholi. 

Sauvage, Bull. Sac. 'pliilom. (7), iv., 1880, p. 228; Passage tie Venus, iii. p. 345 (1885); Vaillant, 
Exped. Antarct. Frangaise, Poiss., p. 22 (1906). 

Depth of body about 6 in the length, length of head 3|- to 4. Diameter of eye 
4f to 5 in length of head, interorbital width about 4, Head mostly naked, with 
granular papillae ; scaly behind the eye and on upper part of operculum. Dorsal 
VI-VII, 24-27. Anal 18-20. Caudal emarginate. Caudal peduncle longer than deep. 
Scales in a longitudinal series 100 to 110 (Sauvage) or 78 (Vaillant), the discrepancy 
probably due to diff'erent methods of counting. Lower lateral line extending forward 
to above middle of anal, its anterior 15 scales overlapj)ed by the upper. Brownish. 

Campbell Island. 

Total length 150 mm. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 279 

4. Dissostichus, Sraitt, 1898. 
Bih. Svensk. Vet.-Akad. Handl., xxiv., iv., No. 5, p. 1. 
Differs from Notothenia in that the teeth are uniserial, spaced, canine-like. 
Patagonia to Graham Land. 

Dissostichus eleginoides. 
Smitt, i.e., p. 2, pi. i. figs. 1-11 ; Vaiilant, Exped. Antard. Fran<;aise, Poiss., p. 36. 

Depth of body about 6 in the length, length of head 3. Diameter of eye about 
5 in the length of head, interorbital width about 4-|^ (? 5^). Maxillary extending to 
below middle of eye ; upper surface of head to nostrils, cheeks and opercles scaly. 
Dorsal IX-X, 27-28. Anal 28-30. Pectoral | the length of head. Caudal truncate 
or slightly emarginate. Caudal peduncle much longer than deep. About 124 scales 
in a longitudinal series ; upper lateral line extending back beyond dorsal, lower extend- 
ing forward nearly to the pectoral. 

Total length of the type, 228 mm. 

Magellan Straits ; Graham Land, 

5. Eleginoj)S, Gill, 1861. 

Proc. Acad. Philad., 1861, p. 522. 

Eleginus (non Fischer), Cuv. and Val., Hist. Nat. Poiss., v. p. 158 (1830); Giinth., Cat. Fish., 
ii. p. 247 (1860). 

This genus differs from Notothenia in the rather small mouth, in the complete 
absence of the lower lateral line, and in the shape of the pectoral fin, 
Chile ; Patagonia ; Falkland Islands. 

Eleginops maclovinus. 

Eleginus maclovinus, Cuv. and Val., I.e. ; Giinth., I.e. 

„ chilensis, Cuv. and Val., o.c., ix. p. 480 (1833); Giinth., I.e. 
Aphritis undulatus, Jenyns, Zool. ''Beagle," Fish., p. 160, pi. xxix. fig. 1 ; Giinth., t.c, p. 243. 

,, porosus, Jenyns, I.e., p. 162; Giinth., I.e. 
Eleginus falklandicus, Richards., "Erebus" and "Terror" Fish., p. 30, pi. xx. figs. 1-3 (1845). 

Depth of body i^ to 5|- in the length, length of head S^ to 4. Diameter of eye 
5 to 8 in the length of head, interorbital width 3 to 5. Maxillary just reaching 
vertical from anterior margin of eye in the young, but not in the adult ; occiput, inter- 
orbital region, cheeks, and opercles scaly ; 14 or 15 gill-rakers on lower part of anterior 
arch. Dorsal VIII-JX, 24-26. Anal 22-24. Pectoral obliquely truncated, with the 
upper rays longest, nearly as long as head. Caudal truncate or emarginate. About 
60 scales in a lateral longitudinal series, and 65 in the lateral line, which nearly reaches 
the caudal fin. Body often spotted or marbled. 

Chile ; Patagonia ; Falkland Islands. 

Here described from several specimens, 120 to 450 mm. in total length. 



280 MR C. TATE REGAN ON THE 

6. Artedidraco, Lonnberg, 1905. 
Swedish South Polar Exped., Fish., p. 39. 
Differs from Notoihenia in the naked body, the absence of the lower lateral line, 
the presence of a mental barbel, the broad union of the gill-membranes to the isthmus, 
and the hooked operculum. 

Coasts of the Antarctic Continent and South Georgia. 

(1) Artedidraco mirus. 
Lonnberg, t.c, p. 40, pi. i. fig. 4 and pi. iv. fig. 14. 
Barbel club-shaped in the male. Depth of body 4 in the length, length of head 
2 1 to 2f. Diameter of eye 3^ to 4 in the length of head ; interorbital width 7 2 to 8|. 
Dorsal III, 23-24. Anal 17. 
Length of types 40 to 92 mm. 

South Georgia. 

(2) Artedidraco skottshergi. 

Lonnberg, t.c, p. 48, pi. ii. fig. 7, pi. iv. fig. 15 ; Vaillant, Exped. Antard. Fran^aise, Poiss., 
p. 46 (1906). 

Depth of body 4|- to 5 in the length, length of head 3. Diameter of eye 3 to 3| 

in the length of head, interorbital width 9 or 10. Dorsal III, 24-25. Anal 18-19. 

Graham Land, 125 m. 

Length of type 57 mm. 

(3) Artedidraco shackletoni. 

Waite, Brit. Antarctic Exped., Fish., ]>. 15, pi. ii. (1911). 
Depth of body 4 in the length, length of head about 3. Diameter of eye 3| in the 
length of head, interorbital width 10. Dorsal V, 27. Anal 20. 
Length of type 146 mm. 
Victoria Land ; off Cape Royds, Ross Island, 30 to 80 fathoms. 

7. Harpagifer, Richards., 1844. 
"Erebus" and " Terror" Fish., p. 11. 
Differs from Artedidraco in the absence of the mental barbel and the development 
of the operculum and suboperculum as strong spines. 
Patagonia to Graham Land and Kerguelen. 

Harpagifer bisjnnis. 

Batrachus hispinis {Callionymus hispinus, Forster), Schneid., Bloch's Syst. Ichth., p. 45 (1801). 
Harpagifer bispinis, Richards., t.c, pp. 11, 19, pis. vii. figs. 1-3, xii. figs. 8-9 ; Giinth., Cat. Fish., 

p. 263 (1860); SniiLt, Bihang Svenslc. Vet.-Akad. Handl., xxiv., 1898, iv.. No. 5, p. 17; 

Vaillant, Exped. Antard. Fran^aise, Poiss., p. 44 (1906); Pappenheim, Deutsche Siidpolar- 

Exped., xiii., Zool., v. p. 177 (1912). 
Harpagifer palliolatus, Richards., t.c., p 20, pi. xii. figs. 5-7. 

Dorsal lil-V, 21-25. Anal 16-20. Coloration variable, usually with bars or 

blotches. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 281 

Total length 100 mm. 

Patao-onia ; Magellan Straits ; Graham Land ; Falkland Islands ; South Georgia ; 
South Orkneys ; Marion Islands ; Kerguelen. 

The Scotia examples are from Station 118, Port Stanley, Falkland Islands, and 
Station 325, Scotia Bay, South Orkneys. 

Family 3. Bathydraconid^. 

The depressed head, produced snout, n on -protractile mouth, and absence of the 
spinous dorsal fin distinguish this family externally from the Nototheniidse ; the genera 
with spatulate snout may be distinguished from the Chsenichthyids without a spinous 
dorsal fin by their scaly body. 

The skeleton of Gymnodraco differs from that of Notoihenia in the more depressed 
skull and more produced rostrum, the elongation of the palatine, which loses its lateral 
ethmoid attachment, and the separation of the mesopterygoid and the metapterygoid, 
so that the upper margin of the quadrate is free. The pectoral arch is as in Notothenia. 
There are 49 vertebrae (20 + 29), the prsecaudals with parapophyses behind which 
the long slender epipleurals are inserted, and the feeble ribs attached to the epipleurals 
at some distance from the centra. 

I have ascertained that Bathydraco agrees with Gymnodraco in the structure of 
the palatine and of the pectoral arch, and in the presence of ribs. 

Synopsis of the Genera. 

I. Body scaly ; snout spatulate ; teeth villiform or cardiform, in bands, without 
canines. 

A single lateral line running to or towards middle of base of caudal fin ; body 
completely scaled . . . . . . . 1. Bathydraco. 

Lateral line running near base of dorsal fin ; body completely scaled. 

2. Gerlachea. 
Lateral line running near base of dorsal fin ; scales in scattered groups. 

3. Racovitzaia. 

[I. Body naked; snout pointed; teeth curved, compressed, close-set in a single 

series, with strong anterior canines . . . . .4. Gymnodraco. 

1, Bathydraco, Giinth., 1878. 
Anji. Mag. Nat. Hist. (5), ii. p. 18. 
Body scaly ; a single lateral line, running to or towards middle of base of caudal. 
Snout spatulate ; jaws with small villiform teeth in bands. 
Antarctic, in deep water. 

GiJNTHER has stated 10 branchiostegals for B. antarcticus, but I find only 7. 
DoLLO gives 6 for B. scotiae, but I count 7 in that species also. 



282 MR C. TATE REGAN ON THE 

(l) Bathydraco antarcticus. 
Giiiith., I.e., and " Challenger" Deep-Sea Fish., p. 47, pi. viii. fig. A (1887). 

Elongate, subcylindrical, the depth 9 in the length, length of head 3. Snout 1^ as 
long as diameter of eye, which is 4 in length of head, interorbital width 20. Lower 
jaw projecting ; maxillary reaching vertical from anterior margin of eye ; cheek com- 
pletely scaled; 16 gill-rakers on lower part of anterior arch. Dorsal 36. Anal 31. 
Caudal subtruncate. Pectoral truncated, as long as head without snout, reaching origin 
of anal. About 140 scales in a lateral longitudinal series, about 60 in the lateral line, 
which is complete. Brownish ; fins dusky. 

South-east of Heard Island, 1260 fathoms. 

Here described from the type, 260 mm. in total length. 

(2) Bathydraco macrolepis. 
Bouleng., Nat. Aniardic Exped. Nat. Hist., ii., Fish., p. 4, pi. i. fig. 3 (1907). 

Depth of body 9 in the length, length of head 3. Snout If as long as diameter of 
eye, which is 4^ in the length of head, interorbital width 14. Lower jaw projecting; 
maxillary reaching vertical from anterior margin of eye ; cheeks naked below the sub- 
orbitals ; 11 gill-rakers on lower part of anterior arch. Dorsal 34. Anal 29. Caudal 
subtruncate. Pectoral as long as head behind middle of eye, reaching origin of anal. 
About 90 scales in a lateral longitudinal series, about 55 in the lateral line, which is 
complete. Brownish ; fins dusky. 

South-west of Balleny Islands, 252 fathoms. 

Here described from the type, 210 mm. in total length. In the original description 
the number of gill-rakers was erroneously given as 6, and of dorsal rays as 39 ; the 
latter number is also shown in the figure. 

(3) Bathydraco scotife. (PI. IX. fig. 4.) 
Dollo, Proc. Roy. Soc. Edin., xxvi., 1906, p. 65. 

Depth of body 9 to 10 in the length, length of head 3|^. Snout If as long as eye, 
the diameter of which is 5 in the length of head, interorbital width 12 or 13. Lower 
jaw projecting ; maxillary not reaching the vertical from anterior margin of eye ; cheek 
naked below the suborbitals; 19 to 22 gill-rakers on lower part of anterior arch. 
Dorsal 38. Anal 31. Caudal subtruncate. Pectoral as long as head without snout, 
extending a little beyond origin of anal. About 100 scales in a lateral longitudinal 
series, 36 to 40 in the lateral line, which ends at a distance from the caudal equal to 
\ its own length. 

Two specimens, 133 and 145 mm. in total length, taken by the Scotia at Station 417, 
71° 22' S., 16° 34' W., off Coats Land: depth 1410 fathoms; temperature 31-9° F. ; 
trawl; 18th March 1904. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 283 

2. 6'erZac/iea, Dollo, 1900. 
Bull. Acad. Roy. Belg. Sciences, p. 195. 

Differs from Bathydraco in that the lateral line runs near the base of the dorsal 
fin ; a second short lateral line above base of anal. 
Deep water off Graham Land. 

Gerlachea australis. 
Dollo, i.e., p. 196, and Res. Voy. " Belyica," Foiss., p. 25, pi. ii. fig. 1 and pi. v. fig. 2 (1904). 
Depth of body 8f in the length, length of head 3|-. Snout twice as long as diameter 
of eye, which is 5 in the length of head, interorbital width 11. Maxillary not reaching 
vertical from anterior edge of eye ; cheek fully scaled. Dorsal 47. Anal 35. 
Pectoral f the length of head. Caudal emarginate. 
71" 14' S., 89° 14' W., 246 fathoms. 
Total length 180 mm. 

3. Racovitzaia, Dollo, 1900. 

Bull. Acad. Roy. Belg. Sciences, p. 317. 

Body with scattered groups of scales ; a single lateral line running near base of 
dorsal fin ; an incubatory pouch between pelvic fins and vent ; in other characters 
similar to Gerlachea. 

Deep water off Graham Land. 

Racovitzaia glacialis. 
Dollo, f.c, p. 318, and Res. Voy. " Belgica," Foiss., p. 29, pi. ii. figs. 2, 3, pi. v. fig. 3 (1904). 

Depth of body 12 in the length, length of head 3;^. Diameter of eye 4 in the length 
of head, interorbital width 25. Maxillary not reaching vertical from anterior margin 
of eye. Dorsal 30. Anal 27. 

71° 19' S., 87° 37' W., 237 fathoms. 

Total length 82 mm. 



&' 



4. Gymnodraco, Bouleng., 1902. 
"Southern Cross" Pisces, p. 186. 

Body naked, depressed anteriorly, compressed posteriorly. Head depressed ; snout 
produced, pointed ; jaws with curved compressed teeth, close-set in a single series and 
with large anterior canines, those of the mandible exposed in front of the snout. 
Operculum with a strong spine with a hooked branch ; suboperculum with a short 
spine ; 6 branchiostegals ; gill-membranes forming a fold across isthmus. Two 
lateral lines. 

Coasts of the Antarctic Continent. 



284 MR C. TATE REGAN ON THE 

Gymnodraco acuticeps. 

Rouleng., I.e., pi. xvii. ; Pappenheini, Deutsche Siidpolar-Exped., xiii., Zool, v. p. 176, pi. ix. tig. 4 (1912). 

Depth of body 8 in the length, length of head about 3. Snout as long as post- 
orbital part of head. Diameter of eye 5 to 6 in length of head, interorbital width 6 to 
7. Lower jaw strongly projecting ; maxillary extending to below anterior margin of 
eye ; gill-rakers short, sometimes almost vestigial except near the angle. Dorsal 28-30. 
Anal 24-26. Pectoral truncated, | as long as head. Caudal truncate. Large dark 
spots on head and body ; fins dusky. 

Victoria Land ; Wilhelm Land. 

Here described from the types, 200 to 300 mm. in total length, from Cape Adare, 
4 to 8 fathoms. 

Family 4. Ch^nichthyid^. 

This family differs externally from the Nototheniidae in the naked body, produced 
spatulate snout, and non-protractile mouth. The skeleton of Champsocephalus esox 
shows several peculiarities. The skull is depressed, with the long rostral lamina formed 
by the frontals and the ethmo-vomer ; the parasphenoid meets the frontals between the 
small lateral ethmoids at the anterior margin of the orbit. The palatine is represented 
by the maxillary process, attached to the lateral edge of the rostral lamina near its 
anterior end, and by a posterior portion articulating with the lateral ethmoid, these 
being connected by a long and slender ligament ; the pterygoid is slender, and there 
is no mesopterygoid. The prseorbital is large, but the suborbitals are unossified. The 
pectoral arch is as in Notothenia. There are 57 vertebrae (28 -(- 29) ; the epipleurals 
are sessile, but the ribs are not ossified. 

I have ascertained that Chmnichthys and Pagetopsis are essentially similar in the 
structure of the rostrum, the palato-pterygoids and the pectoral arch, and in the 
absence of ribs. In all the genera the mouth is verv distensible and the dentaries 
are freely movable on the articulars ; this is the case, to a certain extent, in the 
Bathydraconidse also. 

Synopsis of the Genei'a. 

L Two lateral lines ; pelvic fin-rays all branched or bifid, the middle ones the longest. 

A. Lateral line without bony plates ; a spinous dorsal fin, subcontinuous 

with the soft dorsal. 
No spine on snout ; spinous dorsal of 9 or 10 spines, not more than ^ the length 

of soft dorsal . . . . . . . L Champsocephalus. 

A median spine near end of snout ; spinous dorsal of 12 to 15 spines, more than 

\ as long as soft dorsal . . . . . .2. Pagetopsis. 

B. Lateral line with bony plates. 

A spinous dorsal fin ........ 3. Cliseniclithys. 

No spinous dorsal fin . . . . . . .4. Paraclixniehtliys. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 285 

11. Two lateral lines ; outer rays of pelvic fins longest ; dorsal fins separated by 
an interspace ; lateral line without bony plates . 5. Chsenocephalus. 

III. Three lateral lines. 

Pelvic fins of moderate length 6. Chionodraco. 

Pelvic fins elongate, the rays simple 7. Cryodraco. 



1. Champsocejphalus, GilL 1861. 
Proc. Acad. Philad., p. 509. 

Body naked, elongate ; 2 lateral lines, without bony plates. Eye nearly in 
middle of length of head ; no spine on snout. Jaws with rather narrow bands of small 
sharp teeth, forming only 2 series laterally ; lower jaw not projecting. Gill-rakers 
short, but well developed on all the branchial arches, dentigerous, about 20 on lower 
part of anterior arch. Spinous dorsal fin well developed, its base less than \ that of 
the soft dorsal, with which it is almost continuous ; pelvics comparatively short, with 
the rays normally branched, the middle ones the longest. 

Patagonia ; Magellan Straits ; South Georgia. 

(1) Champsocephalus esox. (PI. X. fig. 1.) 
Chasnichthijs esox, Giiiith., Ann. Mag. Nat. Hist. (3), vii., 1861, p. 89. 

Depth of body 7 to 8 in the length, length of head 3 to 3^. Snout a little longer 
than postorbital part of head. Diameter of eye 7 in the length of head, interorbital 
width 4 to 5. Supraorbital edges not raised. Maxillary extending to below anterior 
part or middle of eye. Uppermost opercular spine shorter than and quite distinct from 
the middle one. Dorsal (IX) X, 33-36. Anal 32-35. 

Body with dark cross-bars. 

Patagonia ; Magellan Straits. 

Here described from five specimens, 200 to 300 mm. in total length, including the 
type of the species. 

(2) Champsocephalus gunna^'i. (PL X. fig. 2.) 
Lonnberg, Swedish South Polar Exped., Fish., p. 37 (1905). 
Depth of body 6| in the length, length of head 3 J. Snout as long as postorbital 
part of head. Diameter of eye 5 in the length of head, interorbital width 3|. 
Supraorbital edges not raised. Maxillary extending to below anterior ^ of eye. 
Uppermost and middle opercular spines only free distally, appearing as a single bifid 
spine. Dorsal IX (X), 37-40. Anal 36-38. Plumbeous, with some broad darker 
cross-bars. 

South Georgia. 

Here described from a specimen of 420 ram, 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 2). 37 



286 MR C. TATE REGAN ON THE 

2. Pagetopsis, gen. nov. 

Body naked, moderately elongate ; two lateral lines, without bony plates. Eye 
behind middle of length of head ; an antrorse curved spine near end of snout. Teeth 
in jaws small, sharp, biserial ; lower jaw slightly projecting ; gill-rakers vestigial or 
absent. Spinous dorsal fin well developed, its base more than | that of the soft 
dorsal ; pelvics rather long, the rays bifid or slightly branched. 

Coasts of the Antarctic Continent. 

Pagetopsis macropterus. 

Champsoceplialus macropterus, Bouleiig., Nat. Antarctic Exped. Nat. Hist.., ii., Fish., p. 3, pi. i. • 
Pappenheim, Deutsche Siidpolar-Exped , xiii., Zool., v. p. 174 (1912). 

Depth of body 4 to 5 in the length, length of head 2f to 2f . Snout nearly | the 
length of head. Diameter of eye 5 in the length of head, interorbital width 4. 
Maxillary extending to below anterior \ of eye. 3 or 4 opercular spines, the upper- 
most with an antrorse hook. Dorsal XII-XV, 28-31, Anal 25-27. Pectoral |, 
pelvics f the length of head. Dark spots and vermiculations on head ; irregular 
double cross-bars on body. 

Victoria Land ; Wilhelm Land, 

Here described from the types, 160 to 250 mm. in total length, from the stomach 
of a seal near Cape Armitage, Ross Island, 

3. Chaenichtfiys, Richards., 1844. 
''Erebus" and "Terror" Fish., p. 12; Giintli., Cat. Fish., ii. p. 249 (1860). 
Difiers from Champsocephalus in having a spine on the snout, the teeth in 
broader bands, and in the bony plates of the lateral line. Dorsal fins separated by an 
interspace. Gill-rakers short, dentigerous. 
Kerguelen. 

(l) Chsenichthys rhinoceratus. 

Richards., t.c, p. 13, pi. vi. ; Giinth., Cat. Fish., ii. p. 249 (1860); Pappenheim, Deutsche 
Siidpolar-Exped., xiii., Zool., v. p. 193. 

Depth of body 6 in the length, length of head 2f. Snout nearly | the length 
of head. Diameter of eye 5J to 6|^ in the length of head, interorbital width 5 to 5^. 
Maxillary extending beyond middle of eye (adult). Head moderately rugose ; supra- 
orbital edges slightly raised. Dorsal VU, 33-34. Anal 30-33. Second and third 
rays of spinous dorsal longest, thence decreasing rapidly. 79 to 84 plates in upper 
lateral line ; a few plates on middle of side. Brownish, with darker spots and 
reticulations. 

Description from the type, a specimen of 450 mm. and a second example of 
175 mm., from Kerguelen. 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 287 

(2) Chxnichthys rugosus, sp. n. 

Eye smaller than in C. rhinoceratus, diameter 8 in head. Head rougher and 
supraorbital edges more elevated. Maxillary shorter, not quite reaching middle of 
eye. Dorsal VIII, 30 ; third and fourth spines longest, fifth as long as first. Anal 29. 
62 plates in upper lateral line ; a nearly continuous series of plates on middle of side. 

A specimen of 400 mm. from Kerguelen. 

A stuff'ed example with VIII, 34 dorsal and 30 anal rays, and 72 plates in the 
lateral line, appears to belong to this species. 



4. Parachsenichthys, Bouleng., 1902. 

"Southern Cross" Pisces, p. 176. 

Differs from Chsenichthys in the absence of the spinous dorsal fin. 
South Georgia ; Graham Land. 

Parachsenichthys georgianus. 

Chxnichthys georgianus, Fischer, Jahrh. Hamburg Wiss. Anstalt, ii., 1885, p. 50, pi. i. figs. 1, 2. 
? „ charcoti, Vaillant, Exped. Aniarct. Fran^aise, Poiss., p. 39, fig. 

Maxillary not nearly reaching the vertical from anterior margin of eye. Inter- 
orbital region narrow, its width less than J the diameter of eye. Dorsal 44. 
Anal 32. 

South Georgia ; Graham Land. 

Total length 490 mm. 

It seems probable that the imperfect fish described by Vaillant, from Graham 
Land, belongs to this species. The figure of the upper surface of the head is at first 
sight rather different from Fischer's, but the differences may be due to the expansion 
of the jaws and opercles and the smaller size of the specimen (head 136 as against 
173 mm.). 

5. Chsenocephalus, gen. nov. 

Body naked, elongate ; two lateral lines without distinct bony plates. Eye some- 
what behind middle of head ; a small prominence at anterior end of ethmoid ; jaws with 
small sharp teeth forming rather broad bands, there being several series even at the 
sides ; lower jaw not projecting ; gill-rakers absent except for 3 or 4 very short 
ones below the angle of the first arch. Spinous dorsal fin well developed, its base 
about \ that of the soft dorsal, from which it is separated by an interspace ; pelvics 
comparatively short, with the two outer rays the longest, enveloped in thick skin, 
but bifid, the others normally branched. 

South Georgia. 



288 MR C. TATE REGAN ON THE 

Chwnocephalus aceratus. (PL XL) 

Cheenichthys aceratus, Lbnnberg, Kungl. Svensk. Vet.-AJcad. Handl., xl., 1906, No. 5, p. 97. 

Depth of body 5 to 6 in the length, length of head 2J to 2|. Snout a little less 
than J the length of head. Diameter of eye 5 to 6 in the length of head, interorbital 
width about 5. Supraorbital edges raised ; operculum with 3 radiating ridges ending 
in spines, the uppermost bifid. Maxillary extending to below middle of eye or beyond. 
Dorsal VII-VIII, 39-40 ; third spine longest, ^ to more than f the length of head. 
Anal 37-38. Pectoral and pelvic fins subequal in length, nearly | the length of head. 
Greyish, with 4 or 5 dark cross bands, the first from spinous dorsal through base of 
pectoral, the second downwards from origin of soft dorsal, the others less regular and 
sometimes with narrower bars developed between them. 

South Georgia. 

Four specimens, 480 to 530 mm. in total length, collected by Mr David Ferguson, 
and presented to the Scottish Oceanographical Laboratory by Messrs Salvesen. 

6. Chionodraco, Lonnbers;, 1906. 

Kujifjl. Svensk. Vet.-AJcad. Handl., xl., No. 5, p. 99. 

Apparently intermediate between Chsenocephalus and Cryodraco, resembling the 

former in fin-structure, the latter in the three lateral lines and the well-developed 

rostral spine. 

Graham Land. 

Chionodraco hamatus. 

Chsenichthys rhinoceratus subsp. hamatus, Lbnnberg, Swedish South Polar Exped., Fish., p. 47 (1905). 
Chionodraco hamatus, Lonnberg, Kungl. Svensk. Vet.-Akad. Handl., xl., 1906, No. 5, p. 99. 

Head 3 in total length (with caudal). Snout nearly ^ length of head, nearly twice 

diameter of eye, and If interorbital width. Dorsal VII, 37. Anal 33. 

Snow Hill. 

Total length 330 mm. 

7. Cryodraco, Dollo, 1900.* 
Bull. Acad. Roy. Belg. Sciences, p. 129. 
Differs from Chsenocephalus especially in the structure of the pelvic fins, with the 
rays simple, the two outer enlarged and prolonged, and in the presence of an additional 
lateral line at the base of the anal fin. 
Graham Land ; Wilhelm Land. 

(l) Cryodraco antarcticus. 
Dollo, t.c, p. 130, and Bes. Voy. " Belgica," Poiss., p. 20, pi. i. pi. v. fig. 7 (1904). 
Depth of body 8 in the length, length of head 3^. Snout 2, eye 4, interorbital width 
5 in the length of head. Dorsal III, 44. Anal 43. Pelvic fin more than l the length 

* A fish from Wilhelm Land, 69 mm. long, is recorded by Pappenheim nnder the name Pagetodes antarcticus. The 
number of fin-rays (D IV, 31 . A 31 ) scarcely justifies this determination, and the fish may well belong to an undescribed 
species. But iis it is so juvenile and even its generic position uncertain, I refrain from giving it a specific name, 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 289 

of the fish, extending nearly to end of dorsal fin. Body with 7 dark transverse 

bands. 

71° 18' S., 88° 2' W., 450 metres. 
Total length 200 mm. 

(2) Gryodraco pappenheimi, sp. n. 

Pagetodes * antardicus (non Dollo), Pappenheim, Deutsche Siidpolar-Exped., xiii., Zool., v. p. 175. 

Leno-th of head 2|- in the length of the fish. Snout 2 in the length of head, diameter 
of eye 5, interorbital width 4. Dorsal V, 45. Anal 39. Pelvics only reaching four- 
teenth ray of dorsal (the prolonged rays perhaps not entire). 

Wilhelra Land. 

Length of the type, 168 mm. to base of caudal. This species is known to me only 
from Dr Pappenheim's description and from some notes and measurements that he has 
kindly sent me. Some of these may be given for comparison with those of the type 
of C. antardicus. The measurements are in millimetres. 

T 1.1, i T> Head to End of j , ,.. , 

Length to Base Opercular Bony Snout. Eye. ^"^^'^'^^''^ 

of Caudal. i-pj^j^ Operculum. Width. 

C. antardicus ... 173 ?56 53 26-5 13-25 10-6 f 

C. pappenheimi . . . 168 68 ?64 32 13 16 



IV. The Systematic Position and Geographical Distribution of the 
Galaxiid^ and Haplochitonid^. 

The Galaxiidse and Haplochitonidse are Teleostean fishes of the order Isospondyli, 
that is to say, they are malacopterous physostomes with a truly homocercal caudal fin, 
with abdominal pelvic fins, and with ribs inserted on autogenous parapophyses. In 
this order the name Salmonoid may be given to a group of fishes with an adipose fin 
usually present, with one supramaxillary or none, with parietals well developed, and 
with oviducts absent or incomplete. The relations of the Salmonoid families are 
indicated in the following synopsis : — 

I. An orbitosphenoid ; an opisthotic ; a mesocoracoid ; vertebrae upturned at base 

of caudal fin 1. SalmonidaB. 

II. An orbitosphenoid ; no opisthotic ; no upturned vertebrae ; meso-pterygoids 
toothless. 
A mesocoracoid ; parapophyses inferior .... 2. Argentinidae. 
No mesocoracoid ; parapophyses lateral . . . .3. Microstomida3. 

* The fish named Pagetodes by Richardson (" Erebus " and " Terror " Fish., p. 15, pi. viii. fig. 3) may have belonged 
to the genus Gryodraco, but in the form of the body, the length of the pelvic fins, and the continuous dorsals it shows 
more resemblance to Fagetoims. Until Richardson's species is rediscovered, the name Pagetodes cannot be used. 

t DoLLo's figure of the upper surface of the head is enlarged, the length of the head, to the end of the bony 
operculum, to 80 mm. and the interorbital width to 16 mm. 



290 MR C. TATE REGAN ON THE 

III. No orbitosphenoid ; no opisthotic ; no upturned vertebrae ; mesopterygoids 
toothed (absent in Salangidee). 

A. A mesocoracoid ; maxillaries dentigerous, entering gape . 4. Osmeridw. 

B. No mesocoracoid ; maxillaries dentigerous, entering gape. 

Head compressed ; mesopterygoid well developed, dentigerous ; ribs ossified. 

5. RetropinnatidsB. 
Head strongly depressed ; no mesopterygoid ; ribs not ossified 6. Salaiigidae. 

C. No mesocoracoid ; maxillaries toothless, behind praemaxillaries. 

Prsemaxillaries not extending whole length of maxillaries ; roof of myodome 
unossified ; no adipose fin . . . . . . 7. Galaxiidie. 

Praemaxillaries nearly reaching extremities of maxillaries ; roof of myodome 
ossified ; an adipose fin . . . . . 8. Haplochitonidse. 

The Argentinidse and Microstomidse are inhabitants of rather deep water, but the 
rest are littoral fishes, many of them entering fresh water and often forming colonies, 
races, or species confined to fresh water. 

It is of some interest to note that the Galaxiidse and Haplochitonidee are related to, 
but more specialised than, the Osmeridse, or Smelt family, of northern seas, Retropinna, 
from the coasts and rivers of Australia and New Zealand, is still nearer to the Galaxiidse 
and Haplochitonidse ; both these families occur in Australia, Tasmania, New Zealand, 
South America, and the Falkland Islands, and there are even two species of Galaxias 
at the Cape of Good Hope. All the species enter fresh water, and the majority seem 
to be strictly fluviatile or lacustrine, but in a few cases species of Galaxias have been 
observed in the sea. 

In 1906 {Proc. Zool. Soc, 1905, ii. pp. 363-384, pis. x.-xiii.) I published a 
revision of the Galaxiidse, and then wrote : — 

"The occurrence of Galaxias raaculatus in the sea has been recorded by Valen- 
ciennes and by Philippi, off the Falkland Islands and off the coast of Chile respectively. 
The observations of Johnston in Tasmania and of Hutton and Clarke in New Zealand 
are to the effect that Galaxias attenuatus descends to the sea periodically to spawn. 
Mr EuPERT Vallentin has seen shoals of little fishes, which I identify with the 
Galaxias gracillimus of Canestrini, in the sea at the Falkland Islands. Eecently 
Galaxias hrevipinnis also has been found to be marine, G. hollansi, described by 
Hutton from the Auckland Islands, proving to be identical with this species." 

Waite [Suhantarctic Islands of New Zealand, p. 586) has recently shown that 
Htjtton's conclusion as to the marine habit of G. hrevipinnis was probably incorrect. 
Eigenmann {Rep. Princeton Exped. Patagonia, iii., Zool., 1909, p. 274) says of G. 
gracillimus : " This is undoubtedly the young of attenuatus " ; and if this opinion, which 
does not appear to be the result of an examination of specimens, be accepted, the known 
marine species of Galaxias would be reduced to two only. 

In my revision I distinguished G. gracillimus from G. attenuatus by the more 



ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 291 

slender form, the smaller head, etc. My specimens, 53-55 mm. in total length, were of 
the same size as the smallest examples of G. attenuatus, but, bearing in mind the extra- 
ordinary larval history of Anguilla, Alhula, etc., I wrote : " Possibly this species may 
be based on a larval form of G. attenuatus, but if so it is remarkable that it has been 
recorded only from South America, and that larval forms of other species have not been 
described." A series of Galaxias attenuatus from the Falkland Islands, since received 
from Mr Vallentin, includes specimens of 20 to 30 mm. which agree with those of 55 to 
60 mm. in form, size of head, etc., and show pretty conclusively that G. gracillimus does 
not represent a stage in the life-history of this species. Mr Vallentin's collection also 
includes some young examples of G. smithii, hitherto known only from the type from 
Sir Andrew Smith's collection ; these are yellowish, with numerous brownish irregular 
vertical stripes. 

The South American species of Galaxias are seven in number, viz. : — 

1. Galaxias attenuatus, Jenyns, 1 842. 
S.E. Australia ; Tasmania ; New Zealand and neighbouring islands ; Chile ; Pata- 
gonia ; Tierra del Fuego ; Falkland Islands. 

2. Galaxias gracillimus, Canestrini, 1864. 
Chile ; Falkland Islands. 

3. Galaxias maculatus, Jenyns, 1842. 
Chile ; Patagonia ; Tierra del Fuego ; Falkland Islands. 

4. Galaxias alpinus, Jenyns, 1842. 
Alpine lakes of Tierra del Fuego. 

5. Galaxias bullocki, Eegan, 1908. 

A?in. Mag. Nat. Hist. (8), i. p. 372. 
Temuco, Chile. 

6. Galaxias platei,^tQ\\idL., 1897. 

Galaxias titcombi, Everm. and Kendall, Proc. U.S. Nat. Mus., xxxi., 1907, p. 92, fig. 
Patagonia ; Argentina. 

7. Galaxias smithii, Eegan, 1906. 
Falkland Islands. 

It should be noted that only the marine species occur both at the Falkland Islands 
and on the continent of South America, and there can be little doubt that Haplochiton 
zebra, with this distribution, will prove to be marine. 

The conclusion that the Galaxiidse are originally marine and are establishing them- 
selves in fresh water is strengthened by their relationship to the Osmeridse; their distribu- 
tion has little bearing on the question of a former extension of the Antarctic Continent. 

The expense of the publication of this Memoir is defrayed from the Government 
Publication Grant administered by the Royal Society of London. 



292 ANTARCTIC FISHES OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 



LIST OF THE PLATES. 



Plate I. 
Raia magellanica x J. 

Plate II. 

Fig. L Chalinura ferrieri. 
Fig. 2. ,, whitsoni. 

Plate III. 

Fig. I . Cynomacrurus piriei. 
Fig. 2. Neohythiies bi-ucei. 

Plate IV. 

Fig. I. Bovichtliys angustifrons. 
Fig. 2. Cottoperca macrophthalma. 
Fig. 3. „ gobio. 

Plate V. 

Fig. \. Austrolycus depressiceps. 

Fig. 2. Cottoperca maerophthalma x if. 

Plate VI. 

Fig. 1. Gxaioperca coatsii. 

Fig. 2. Notothenia trigramma x f. 







Plate VII. 


Fig. 
Fig. 


1. 

2. 


Notothenia ramsayi. 
„ loiltoni. 

Plate VIII. 


Fig. 
Fig. 
Fig. 
Fig. 
Fig. 


1. 

2. 
3. 
4. 
5. 


Notothenia angustifrons. 

„ marionensis. 

„ acuta X 1|. 
Trematomus loennbergii. 
Synaphobranchus australis 

Plate IX. 



Fig. 1. Bovichthys decipiens x 1|. 
Fig. 2. Bathylagus glacialis. 
Fig. 3. Lycenchelys antarcticus. 
Fig. 4. Bathydraco scotix. 
Fig. 5. Bovichthys diacanthus. 

Plate X. 
Fig. 1. Ghampsocephalus esox. 



Fig. 2. 

Plate XL 

Chasnocephalus aceratus. 



gunnari x -^. 



Trans.Roy Soc Edii? 



Regan: "Scotia" Antarctic Fishes. Pl. I 



VoLXLJX. 






■ ft . . < 




■^*»SSi,' ' *S-<^' 



r 







^' I x'. 




^..H. SeaxU,ael.et.litk. 



RAIA- L/LAGELLANICA. 



IHAJ.tk| UD.p . 



^'J 



^^^-^'-5'^. .. ^^^. 



r- 



*"V, 



7--^ 







A > - ' / 

i^ff^l 'i I ii / 











^i 



'r:^A 




o 

CO 

w 
X 

pq 
o 



M 
M 
H 

CD 

o 

<^ 

o 
o 



ran 



.Roy.Soc. EdinT^^^^^^ "Scotia'^ Aijtarctic Pishes, Pltv. 



VoIXLIK. 




arla.del.et litK. 



1, BOVICHT.HYS AN GU S TIF RONS . 
2. COTTOPERCA MACROPHTHALMA . 3. C. GOBIO. 



Kutli imp. 







o 

(^ 

o 



p:; 

O 

o 
o 

03 



CQ 

W 

o 

H 
CO 
CO 

ft 

CQ 

o 

o 

M 

CO 




n^ 








W 

B 
O 

EH 

O 

!3 

CO 



M 
CO 

o 
o 

o 
o 

M 



o 



^^,v.-.-..s,^^ 




ras.Roy. SocEdin? ^^q^. "Scotia" Antarctic Fishes, Pl.k. 



Vol.XLTX. 



,^P 




2. 






3a. 









^a. 




'C*-^ 




5a. 

A.I|3ei.rle del et lith. 

1. BOVICHTHYS DECIPIENS 2.BATHYLAGUS GLACIALIS. 
3. LYCENCHELYS ANTARCTICUS 4. BATHYDRACO SCOTIAE. 
5. BOVICHTHYS DIACANTHUS. 



Huttjimp . 



<^ ^'^S^ 






"A 

6 

ai 



N 
o 

CO 
CO 

w 

W 
o 
o 

CO 



O 

r-i 



;! 










W 
O 
< 

iJ 
<: 
W 
ft 

o 
o 



o 



l^ wuse^r. 




( 293 ) 



II[ A Monograph on the general Morphology of the Myxinoid Fishes, based on 

a study of Myxine. Part V. The Anatomy of the Gut and its Appendages. 
By F. J. Cole, D.Sc. Oxon., Professor of Zoology, University College, Reading. 
Communicated by Professor W. A. Herdman. F.R.S. (With Four Plates.) 

(MS. received October 5, 1912. Read December 16, 1912. Issued separately June 9, 1913.) 



CONTENTS. 



PAGE 

A. Habits 294 

B. Nasal Aperture and Mouth .... 295 

C. Body Cavity 296 

D. Respiratory Organs 298 

1. Structure of the Gills and Gill Ducts . 298 

2. Variations in the Branchial Organs . 303 

E. The Mucous Surfaces 304 

F. Histology and Divisions of the Gut . . . 308 

G. Liver and Biliary App>aratus — 

1. Superficial Anatomy . . . .317 



2. General Anatomy of the Liver 

3. Bio-chemistry of the Liver 

4. Histology of Gall Bladder 

5. Histology of Hepatic Ducts . 

6. Histology of Bile Duct . 



PAGE 

320 
324 
326 
327 

328 



7. "Pancreas" 329 

H. Thyroid Gland 332 

J. Cloaca 335 

K. Literature ....... 338 

L. Explanation of Plates 341 



The first four parts of this work, on the skeleton, muscles, and vascular system, 
were published in the Transactions of the Society in 1905, 1907, 1909, and 1912. 

The present section does not include the teeth, which, belonging properly to the 
skin, will be described when the skin and its appendages are dealt with. It will be 
noticed from time to time that I have to record striking differences from results 
obtained by other authors. Such differences must, of course, be recorded ; but in 
recording them it must not be understood, unless the contrary is expressly stated, that 
the observations under discussion are necessarily inaccurate. One cannot investigate 
Myxinoid anatomy without frequently coming up against somewhat astonishing varia- 
tions, and 1 have long been convinced that there must be races or colonies of Myxine 
which have not been distinguished specifically, but which, nevertheless, possess 
anatomical features in common. To give one instance out of many. Mr R. H. Burne 
has described in detail an anal slime gland in Myxine. In my sections of examples up 
to 25 cm. there is no trace of this structure. On examining Mr Burne's sections I 
find that the dorsal chamber of the cloaca has been converted into what anyone would 
interpret as a slime gland, similar to the series at the side of the body developed in 
connection with the skin. Mr Burne, therefore, is both right and wrong at the same 
time, but no one with the material at his disposal could have deduced what the 
supposed anal slime sac really was. 

I must express my indebtedness to Professor A. Meek and his assistant, Mr B. 
Storrow, for their very successful efforts to obtain living specimens of Myxine, and for 
the hospitality of the excellent marine laboratory at Cullercoats. The experiments on the 
bio-chemistry of the liver were carried out by my museum assistant, Mr A. H. Malpas. 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART IL (NO. 3). 38 



294 PROFESSOR FRANK J. COLE 

The expenses of this research have been defrayed by a grant from the Government 
Grant Committee. 

A. Habits of Myxine. 

The predaceous habits of Myxine have been known for a long time. Linn^us, who 
classified it among the "vermes intestina," says: "Habitat in Oceano europseo, pisces 
intrans, devorans, aquam in gluten vertens." Pennant, in 1766, refers to the fish on 
the lines being reduced to skin and bone by Myxine. 

At Cullercoats Myxine is found on a muddy bottom, not as a rule inside the 
23-fathom line, and common at 25 fathoms. It also occurs on a clayey bottom from 
25 to 50 fathoms, but is not so numerous. On their own ground they must be as 
plentiful as earthworms. They will migrate on to an artificial muddy bottom, such 
as when a quantity of dredged mud has been dropped by hoppers on to an originally 
hard bottom, and in one such case they were known to have moved inshore from 6 
to 3 miles. 

All fishermen agree that the Hag enters the line fish by the gills and not by the 
mouth, and completely cleans it out, so that when the hooked fish is hauled up it is 
simply a bag of skin and bone. This was described by Fleming in 1823, and Meynell 
mentions that 123 Myxine were taken out of one codfish at Redcar in 1843. At one 
time Myxine were so common as to constitute a serious menace to the line fishery, but 
the fishermen allege that since trawling has developed, the Hags have largely dis- 
appeared, due doubtless, if correct, to the trawlers reducing the food of the Hags. It 
is further asserted by the fishermen that Myxine will not touch a fish that has been 
long dead, but that they only attack dying or newly dead line fish. They are readily 
captured on hooks baited with the foot of the limpet and salt herring. 

The Hag swims freely and easily, like an eel, by lateral undulations. It can swim 
backwards, and usually escapes from a bucket tail first. Its power of secreting slime 
has been greatly exaggerated, unless the extensive experience which I have now had 
is greatly at fault. Goode and Bean state : "A single Hag will fill a two-gallon 
bucket with slime mingled with water in a few seconds, and after a slight interval can 
repeat the operation with ease." I have seen nothing in any sense approaching this 
performance. 

Owing to the fact that hooked Myxine are usually damaged by the hook — in some 
cases the hook may be found as far back as the anus — I devised the following method 
of capturing large numbers of Hags in 1903. A line about a quarter of a mile long 
with, say, 35 black jack [Gadus virens) tied to it at intervals was anchored at each end 
to the sea bottom in 25 fathoms of water. From the anchor at each end a line with 
a fioat went to the surface, so that the position of the bottom line could be located. 
The bait must not be left down too long — in three hours, for example, all the black jack 
would be destroyed and the Hags gone. In this and other experiments I was able 
to establish : — 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 295 

1. That Myxine never attacked living fish on the lines, but preferred fish recently- 
living to stale fish. They did, indeed, attack the latter, thus disproving the assertion 
of the fishermen. I agree, however, with the fishermen that Hags never attack living 
free swimming fish, and are therefore not parasites. 

2. As indicated by the slime, the Hags enter the fish by the gills rather than 
by the mouth. This, doubtless, to escape the teeth. When the fish were hauled up the 
Hags could be seen sticking out of them in all directions, and great numbers were 
lost by escape through the mouth. Only a few came out through the gills. 

3. The body cavity was usually entered dorso- laterally. The liver was eaten first, 
then the gut and heart, and finally the flesh between the skin and backbone was 
attacked at the posterior end of the body cavity, the Hags working forwards until 
the dorsal muscles were entirely devoured. The whole operation takes about two 
hours. 

4. In a dying or feeble fish the Hags stop the action of the gill cover by blocking 
it with slime. As soon as the motion of the operculum ceases, they enter the body 
through the gills. 

5. It was invariably noticed that in the region of the back the Hags did not 
touch the spinal nerves, which looked as if they had been roughly dissected out. This, 
t believe, is not due to any preference for muscle, but simply owing to the peculiar 
character and action of the dental apparatus, which would rasp away the muscle and 
leave the stringy nerves. 

The respiratory current in Myxine is easily demonstrated in the living animal by 
placing a few grains of eosin in front of the nasal aperture. Almost immediately 
water discoloured by eosin will be seen issuing from the branchial apertures. The 
respiratory current is a constant and steady stream, and there is only occasionally 
a sharp discharge from the gills. 

Myxine can evidently distinguish light from darkness. When a number were 
placed in a large tank with one dark end, all the more active and healthy fish 
migrated to the dark end, and invariably returned there when brought back. 

B. Nasal Aperture and Mouth. 

The nasal aperture is relatively very large, as in PalsBospondylus, and is median, 
dorsal and terminal. It is compressed from side to side, and is overhung above by a 
prominent forwardly projecting lip-like process. On each side of the latter is a short 
anteriorly projecting pointed tentacle for anterior touch, and below and in front of it, 
but at the side of the nasal aperture, is another and similar tentacle on each side. 
In the living animal the latter tentacle juts out almost at right angles to the long 
axis of the body and somewhat downwards, and is for lateral touch. 

The mouth is circular and sub-terminal. There are no obvious jaws. It is 
situated on the ventral surface behind and below the nasal aperture. In both dead 



296 PROFESSOR FRANK J. COLE 

and living specimens it is contracted and puckered at the edges. At the sides of 
the mouth, somewhat in front, is a long slender tentacle (the longest on the head), 
which is directed in a curve downwards and either forwards or backwards. This 
tentacle can be rotated forwards, and is for ventral touch. Internal and posterior to 
this on each side is a smaller tentacle with a very expanded base. From the dorsal 
front border of the mouth in preserved specimens two prominent diminishing ridges 
pass backwards into the mouth. There is no buccal funnel as in Petromyzon. 

C. The Body Cavity. 

When a median ventral incision is made, and the body walls reflected, we note, first 
of all, the peritoneum — a thin, glistening membrane lining the body cavity, and through 
which the muscles of the body wall are plainly visible. Anteriorly in the body cavity 
are seen the two lobes of the liver, one behind the other, the posterior lobe being rarely 
partially subdivided into two, whilst between these two lobes on the right a portion of 
the large gall-bladder projects. The very wide and perfectly straight intestine is 
noticed passing back to the cloaca from over the posterior lobe of the liver. To the right 
of the intestine is usually seen the single gonad {i.e. its ovarian portion), which in the 
individuals predominantly female more or less affects the shape of the intestine accord- 
ing to the time of year. This part of the gonad is formed in a genital duplication of 
the dorsal mesentery (mesoarium) on the right side only (cp. Part II., fig. 1), and thus 
corresponds to the right genital gland. In a female hermaphrodite the ovarian region 
of the gonad will be found to contain a number of very large elongated oval eggs, whilst 
in the male hermaphrodite it may not be seen at all without further examination. 

If the posterior section of the intestine be examined carefully (fig. 5), it will be 
noticed that as it approaches the cloaca it begins to fuse at the mid -ventral line with 
the body wall — thus forming at the beginning of the fusion a rudimentary mid-ventral 
mesentery {mes.'). The fusion of the intestine with the body wall, however, at once 
extends obliquely upwards and backwards [mes."), thus completely obliterating the 
body cavity below the intestine. The body cavity, hence confined to the roof of the 
gut, becomes diminished as it is continued backwards, and opens finally into the cloaca 
by the large, single, median porus genitalis or abdominal pore (p-g.), as elsewhere 
described. 

The intestine is suspended by a median dorsal mesentery (mes., cp. also Part II., 
fig. 1). As shown in fig. 5, this mesentery terminates by a curved border a short 
distance in front of the abdominal pore, and thus the two dorsal halves of the body 
cavity are in free communication for a short region anterior to the genital pore. This 
is obviously necessary in order that the large eggs may escape with facility. 

If the anterior extremity of the body cavity be now examined on the right side, 
and the liver turned over to the left, it will be seen that the pericardial coelom with 
its contained heart is situated over the anterior extremity of the front lobe of the liver, 



J <-K 






ON THE GENERAL MORPHOLOGY OF THE MYXTNOID FISHES. 297 

and also that the latter cavity communicates with the body cavity on this, the right, 
side by a large pericardio-peritoneal foramen. This is the only one present, the 
pericardium being imperforate on the left side. There are, however, two foramina in a 
the Ammocoete, but none in the adult Petromi/zov . In a 34-cm. Hag the pericardio- i^^^'Cr**" '/^^ 
peritoneal foramen was an elongated aperture 9 mm. long when slightly stretched, and / f^^^'^JTe 
was directed obliquely forwards from right to left. Its right wall is formed by the i_ c**"^ ' 
serosa of the gut, and here the very large supra-intestinal or portal vein (fig. 4,^.1;., 
and cp. also Part II., fig. 1), and in some cases even the portal heart {x>Ji.), projects 
boldly into the foramen. It is usually stated that the portal vein passes through the 
foramen on its way to the portal heart, but the vein lies morphologically outside the 
foramen, and is in fact only accidentally related to it. On the other hand, the common 
portal vein {c.p.v.) is neither apparently nor really associated with the foramen. The 
left wall of the latter is formed by a special duplicature of the peritoneum having a 
well-defined and slightly thickened free border. This structure, doubtless, corresponds 
to the Selachian pericardio-peritoneal septum. The presence of such a large pericardio- 
peritoneal foramen in Myxine is possibly correlated with the existence of an important 
pericardial pronephros. 

A little behind the heart the intestine, which has so far been suspended by the 
median dorsal mesentery, takes an upward turn, and therefore the two sheets of the 
peritoneum, instead of becoming opposed to form the median mesentery, pass at once 
separately on to the gut, and from the gut they immediately reach the anterior lobe of 
the liver, instead of forming a median ventral hepatic ligament such as suspends the 
posterior lobe of the liver to the intestine. It is from these lateral sheets of peritoneum 
that the pericardium is derived. There is thus no discontinuity in the mesentery at the 
anterior end of the body cavity as there is at the rectum, apart, of course, from the 
pericardio-peritoneal foramen itself. 

When the intestine is displaced to one side, four large vessels are seen attached to 
the body wall at the mid-dorsal line. The outermost pair are the segmental or 
pronephric ducts ; the innermost, the two posterior cardinal veins. It will be noticed 
that of the two cardinals the right is perceptibly smaller than the left, as in Bdellostoma 
and also in preserved specimens the left is the more usually blocked with blood. The 
dorsal aorta is hidden from view (except anteriorly) between, and dorsal to, the two 
cardinals, and is, as a rule, only exposed by dissection (but this varies). The arteries 
from the aorta to the alimentary canal and gonad (cp. Part II., fig. 1) are seen emero-ing 
between the two cardinals and apparently from them. 

Almost at the mid-dorsal line of the whole length of the gut, and to the right of 
the attachment of the dorsal mesentery, courses the very large portal vein (Part II. 
fig. 1, p.v.). 

The anterior region of the body cavity, with the opening from the abdominal into 
the pericardial coeloin is figured by Goodrich (24, p. 44) in Myxine glutiviosa. 



298 PROFESSOR FRANK J. COLE 

D. The Respiratory Organs (Figs. 2, 6, and 12). 

The coarse anatomy of the breathing apparatus of the Myxinoids has been so often 
described that an acquaintance with it may be taken for granted. Further, the general 
structure is sufficiently indicated in the reconstruction given as fig. 2. The peri- 
branchial or pleural sacs, and the relations of the gills to them, are fully described in 
my Part IV. pp. 220-222. I may, therefore, at once proceed to a more detailed 
account of the anatomy of the parts. 

1. Structure of the Gills. 

Under the serous membrane with its long flat nuclei the outer gill wall exhibits a 
loose connective-tissue layer in which muscle fibres are lodged. They are in two sparse 
layers, an external circular or concentric, the axes being the gill ducts, and an internal 
radial. The concentric fibres do not pass right round the gill, but each bundle of fibres 
forms only a more or less short segment of the circle. The fibres histologically are 
intermediate between the plasmic and aplasmic types described in my second part. 
They are distinctly striated ; the nuclei are large, vesicular, and peripheral ; but there is 
only a slight quantity of sarcosplasm present, most of it being collected round the 
nucleus. The fibres are somewhat flattened, and average about 37 "5 m by 12 m, and 
both the transverse and longitudinal striation show up with diagrammatic clearness 
in material stained with iron hsematoxylin. 

The fundamental ground plan of the Myxinoid gill is quite simple. There is a 
tube connecting the gut with the exterior, which we may call the gill duct. At one 
place the epithelial lining of this duct throws out a number of lamelliform evaginations 
— about 8-10, but the number in the adult is difficult to enumerate owing to secondary 
modifications — the long axes of which are radial and the short axes parallel to the long 
axis of the duct. Hence at this place a considerable swelling of the duct supervenes 
to form the cake- or pouch-like gill. The unmodified proximal and distal portions of 
the duct remain as the afferent and efferent gill ducts respectively (figs. 2, 6, and 12, 
a.g.d., e.g.d.) 

It therefore results in a section at right angles to the long axis of the duct itself 
that we have a star-shaped arrangement. There is a central cavity, the cavity of the 
duct, and a number of compressed radial chambers opening into it. The tissue between 
the walls of contiguous evaginations is highly vascular, and constitutes the respiratory 
apparatus of the gill. This tissue projects inwards from the outer wall or periphery 
of the gill as a number of radial plates, each one clothed by two sheets of epithelium — 
the walls of two adjacent evaginations. A gill lamella, therefore, is formed by the 
contiguous walls of two epithelial evaginations plus the intervening vascular tissue. 

The afferent gill artery (afhr.), on reaching the gill, forms a circle round the 
efferent gill-duct (cp. fig. 6). From this circle a number of radial arteries arise like the 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 299 

spokes of a wheel. These are easily seen in an injected gill coursing quite close to the 
outer or more convex surface of the gill from the centre to the edge. The radial arteries 
in their turn give off a number of vessels at right angles which travel in the core of 
the gill lamella between the epithelial sheets above mentioned, and join up to form, on 
the inner or less convex (sometimes concave) surface of the gill, another series of radial 
or spoke-like arteries, which all open into a further circular vessel round the afferent 
gill duct. From the latter circle the two efferent branchial arteries {ef.br.) arise, and 
behave as described and figured in my Part IV. 

The actual structure of the gill, however, is not as simple as this. The radial epithelial 
evaginations, for example, often dichotomise at least twice, and sometimes five times — 
twice in the lateral regions of the gill {i.e. near the afferent and efferent surfaces), and 
five times in the central region. This produces at the circular margin of the gill a 
large number of diverticula instead of a few. Then the radial vessels connected with 
the rings round the afferent and efferent gill ducts anastomose and branch. Again, the 
respiratory surface of the gill lamellae does not remain simple and smooth. A number 
of ridges or pleats are thrown out which themselves dichotomise several times along 
their long axes so as to produce in transverse section a dendritic appearance — not, 
however, over the whole of the gill lamella, for, axially and especially laterally, i.e. nearer 
the greater gill surfaces, the gill lamella remains smooth. The same result would, of 
course, be produced by a series of ingroivths of the epithelial covering of the lamella, 
and this is perhaps how the dendritic structure seen in transverse section has arisen. 

The ridges or pleats stretch straight across the gill lamella from one surface of the 
gill to the other, at right angles to the greater surfaces of the gill. Hence, they are cut 
longitudinally in transverse and horizontal sections of the gill, and transversely in 
vertical sections. Hence also the dendritic structure is only obvious in vertical sections. 

A vertical section of a well-injected Myxinoid gill is a beautiful but deceptive object. 
One sees what appears to be a number of radial vessels converging inwards from the 
periphery to the centre of the gill. Proximally these vessels are simple in structure, 
but as they approach the centre of the gill they become highly dendritic. Lateral 
branches are given off which bifurcate a number of times. There is apparently no 
question of arteries and veins. Each branch lodges only one vessel, and therefore seems 
to be a vascular cul-de-sac. 

A closer examination and comparison with sections in other planes reveals the 
explanation. Proximally each lamella exhibits in vertical section a linear series of 
vessels cut in transverse section. These are often connected up by vertical anastomoses, 
and this conveys, in vertical section, the wholly false impression of a radial vessel. The 
dendritic structure is again equally misleading. Here also, so far from having vascular 
culs-de-sac, we have in reality a large number of transverse sections of very small vessels 
coursing at right angles to the plane of the section from one large surface of the gill to 
the other, and connected up at right angles by innumerable branching anastomoses. In 
a transverse section these anastomoses show very well as extensive aggregations of 



300 PROFESSOR FRANK J. COLE 

massed dots in the more central portions of the gill. We have, in fact, a dejfinite 
capillary system in the gills, although a somewhat coarse one. Near the circular margin 
of the gill a few respiratory vessels may pass straight across from the afferent to the 
efferent radial vessel without modification, as shown in fig. 6. The type of gill circula- 
tion described above, carried much further, gives us the example of circulation found in 
the gills of the lamprey as described by Favaro.* 

The branches of the afferent radial vessel do not usually pass straight across the gill 
to become the factors of the efferent radial. Sometimes they do, and one can then 
trace across the gill lamella a single vessel coursing from the afferent to the efferent 
vascular side of the gill, and giving off anastomoses (about 50) as it goes along. But 
more frequently the direct continuity is broken, and the branches of the afferent radial 
interdigitate with the factors of the efferent vessel, as shown in fig. 6. 

In section it is easy to distinguish within the gill the branches of the afferent 
branchial artery from the factors of the efferent branchial. The walls of both include a 
few unstriated muscle fibres with long flat nuclei, but the lining of the afferent vessels is 
very characteristic. Here we find, projecting into the cavity of the radial vessels and 
their branches, large cells of varying shape separated from each other by distinct 
intervals. They may project 12 m into the cavity of the artery, and are separated 
by spaces of about 2"5 m, but sometimes by much wider gaps. At their base they 
are about 10 m across. Each cell contains a large vesicular nucleus — sometimes two 
— and the cytoplasm lodges a number of yellow brown particles not coloured by all 
dyes, but staining black with iron hsematoxylin. 

These cells evidently constitute a blood " gland" of some kind. Giglio-Tos regards 
the lymphoid tissue of the valvula spiralis of the Ammocoete as the source of both red 
and white corpuscles of the blood ; but this view is attacked by AscoLi, who regards the 
circulating blood of the Ammocoete as the formative sphere of the red corpuscles, as in 
the embyro, and finds, as K. B. Shreiner has also done in Myxifie, the new corpuscles 
dividing in the blood. Maas admits his inability to trace the source of the red 
corpuscles of Myxine, and was not able to observe any early stages in the blood. He 
states, however, that the body of these corpuscles, stained with iron ha^matoxylin, 
exhibits numerous dark small accretions. I hope to be able to show in my next part 
that the lining cells of the afferent vessels of the gills described above are the source of 
the red corpuscles of the blood. 

Besides the afferent and efferent arteries the connective tissue of the gill lodges a 
large number of irregular spaces, or lymphatics, having an inconspicuous but definite 
lining with flat nuclei. Usually these spaces contain only a few blood corpuscles, but 
they may contain many. They partially fill up in injected material. They are found 
not only in the wall of the gill, but extend also into the coarser portions of the gill 
laraellaj. On the side of the efl'erent gill duct the lymphatics communicate with the 
peribranchial sinus (cp. Part IV, p. 221) by a large but short channel, which enters the 

* Atli Accad. tici. vend, treat, istr., 1905. 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 301 

gill in company with the afferent branchial artery. Similarly, both the efferent branchial 
arteries of a gill are associated with lymph channels placing the lymphatic spaces on that 
side of the gill in communication with the dorsal sinus system (cp. Part IV. pp. 220-1 ), 
but the lymphatics are not so well developed on this side of the gill. 

The large mucous cells, which are so characteristic of the Myxinoid gill gut do not 
occur within the gills themselves. In one case, however, 1 found numerous mucous 
cells of the glassy type in the thicker epithelium on the efferent gill-duct side of 
the gill. 

The whole of the complex cavity of the gill is lined by epithelium, which is better 
developed near the two great surfaces of the gill, where the gill lamellae are simple in 
structure, and which, further, is stronger on the afferent arterial side than the efferent. 
There is roughly an extra rank of nuclei on the afferent side, and in some specimens 
the disproportion is greater than this. The free surface in places is bounded by 
flattened squamous cells, and under these there may be up to four irregular ranks of 
nuclei, some of which here and there are observed to be undergoing mitosis. The cell 
boundaries can be easily distinguished under the higher powers of the microscope, the 
epithelium then having in vertical section the appearance of a mosaic. 

In the complex or more respiratory portions of the gill lamella the blood is only 
separated from the water by what appears to be, in longitudinal sections of the vessel, 
a single row of shallow cells about 2 /a thick, and having flattened nuclei up to 10m 
long, separated by intervals of 20 m- In transverse sections of the vessels these 
nuclei are seen to follow the curves of the vessel, and are hence markedly crescentic. 
They measure about 10 m this way also, and are therefore circular concavo-convex 
discs. Under the highest powers of the microscope, however, this boundary separating 
water from blood appears as a very thin, almost structureless, membrane having definite 
outer and inner borders, the nuclei above described being situated on the vascular side, 
and internal to the inner border. The above statements apply only to that part of 
the vessel which projects or bulges into the cavity of the gill, i.e. that portion of it in 
direct contact with water. Elsewhere the nuclei are of a different character, and are 
embedded in the wall of the vessel. 

Although appearances very often suggest it, I have not satisfied myself of the 
existence of intra-cellular capillaries in the Myxinoid gill. 

The efferent gill duct is lined internally by a layer of mucosa thrown into about 
twelve conspicuous folds. Underneath this is a somewhat extensive vascular zone of 
fibrous connective tissue forming the submucosa. This in its turn is surrounded by 
the irregular but well-developed muscular coat of striated fibres, of which the external 
fibres are mostly circular and the internal ones longitudinal. External to the muscular 
coat we have no definite boundary, but some very loose fatty connective tissue. The 
lining epithelium is very similar to that of the gill itself : there are about three ranks 
of nuclei ; cell boundaries are easily distinguishable in methyl-blue-eosin preparations — 
the superficial row of cells being flattened, the cytoplasm vacuolated, with a definite 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 3). 39 



302 PROFESSOR FRANK J. COLE 

free border. Occasional mitoses are seen, but there are no gland or mucous cells. A 
few of the glassy mucous cells were, however, seen in the afferent gill duct, and Schreiner 
says that there are "by far not as many slime cells in the afferent gill ducts as in the 
mouth." The muscle fibres are, on the whole, similar to those of the gills, except that 
they are coarser, and the striation is not so distinct. 

The external branchial apertures are conspicuous openings situated far back near 
the middle line on the ventral surface, just in front of the anterior extremity of the 
pre-anal " fin," and a little behind the anterior third of the body. They are always 
asymmetrical as regards size, but not invariably so as regards position. The right is 
always the smaller of the two (and naturally so, as the ductus oesophago-cutaneus only 
occurs on the left side), and it is situated a little in front of the left ; externally the left 
opening is sometimes partially, but rarely completely, divided into two — in the latter 
case the ductus oesophago-cutaneus having a separate external aperture. In 1815 Sir 
EvERARD Home published a figure by William Clift in which three such openings are 
indicated. J. Muller, however, states that he has never seen this, but Maas finds in 
his youngest specimens the branchial cloaca and the ductus oesophago-cutaneus entirely 
separate, so that only their openings on to the outer skin come together, like a pair of 
spectacles. He supposes therefore that the common duct is only formed with age, bur 
adds that the ductus never opens medially, but always on the left. There is no doubt 
in fact that the ductus is a potential left gill. When the ductus opens separately the 
two apertures are not side by side, as in Clift's figure, but antero-posterior, the smaller 
anterior one being the orifice of the branchial cloaca of its side, and the larger posterior 
one that of the ductus. Internally, of course, the two structures are always distinct. 

The structure of the gill of Bdellostoma has been described by Jackson (30). He 
has evidently, however, directed but little attention to this section of his work, and his 
scheme of the circulation, representing, as it does, one side of the gill as arterial and the 
other as venous, is manifestly unsound even on a priori grounds. He has failed here to 
recognise the fundamental fact of the interdigitation of the afferent and efferent vessels 
— a condition absolutely essential if the gill is to be a respiratory organ at all. Again, 
his fig. VIII. is a misinterpretation of the apparent dendritic structure of the gill 
lamella as seen in vertical section. As I have pointed out above, it is fatally easy to 
commit this error, especially if one does not realise how impossible vascular culs-de-suc 
must be in the Chordate gill. 

The development of the Myxinoid gill follows the course we should deduce from a 
knowledge of its anatomy. In Bdellostoma, according to Stockard (52), the gill is 
originally tubular, only the oesophago-cutaneus duct remaining in the tubular stage 
throughout life. The gill tubes are entirely endodermal. At the point where the future 
gill pouch is to develop the lumen of the tube becomes enlarged, and then, by a folding 
of the walls, the characteristic radiate gill is produced as outlined in the anatomical 
description above. 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 303 

2. Variations in the Number and Vascular Supply of the Gills. 

Amongst the many dissections of the gills which I have made, the following departures 
from the normal number of six on each side were encountered : — 

(a) Seven gills on the left side.—The additional one was apparently the last, since it 
was supplied by a twig from the last afferent branchial artery. Normal on the right side. 
In this specimen the ventral aorta was continued forwards as an impaired artery on to 
the club muscle. 

(b) Seven gills on both sides. — On the left side all the gills were of much the same 
size, except that the middle ones were slightly the larger, and there were five afferent 
branchial arteries. The first and last afferent branchials split quite near their origin, to 
supply two gills each. On the right side the gills alternated with those of the left, the 
first being situated between the first and second of the left side, and so on backwards. 
The size of the gills was the same as on the other side, except that the last was very 
small, and only about one-third the dimensions of the sixth. There were six afferent 
branchials on this side, one to each gill, but the last small gill was supplied by a twig 
from the sixth, given off near the place where it entered the sixth gill. 

(c) Seven gills on the left side. — First gill more anterior and rather more dorsal 
than usual, and situated laterally and dorsally to the posterior end of the club muscle, 
not reaching as far back as the latter point. Usually about half the first gill overlaps 
the club muscle, but the forward position above described may occur without any 
variation in the gills. Last or seventh gill much the smallest, and apparently the extra 
gill, as it is supplied by a branch of the last or sixth afferent branchial. Ductus oesophago- 
cutaneus present. Gills normal on the right side. 

(d) Seven gills on the left side. — Similar case in every respect to (c), except 
that the afferent branchial for the seventh gill arose from the aorta side by side 
with the sixth, and was not merely a branch of the latter. The extra gill is again 
the smallest, but the disparity is not so great as in preceding case. Gills normal on 
the right side. 

(e) Seven gills on both sides. — ^On the left the first is in a much more dorsal posi- 
tion than the others, and lies at the side of the club-shaped muscle, which latter projected 
slightly behind the posterior border of the gill. First afferent branchial split to supply 
the first two gills. Apart from the usual decrease in size from before backwards, the 
seventh gill was quite normal, and had its own afferent branchial — well separated from the 
artery supplying the sixth gill. On the right side the first gill does not differ in position 
from the others, but the whole seven form a graduated series of which the last is the 
smallest, but not remarkably small. Each of the seven gills was supplied by an afferent 
branchial artery arising separately from the aorta. 

(/) Seven gills on the left side. — First rather more dorsal than the others, and lies 
at the side of the club muscle, which projects slightly behind it. First afferent 
branchial splits into two to supply the first two gills. Last gill distinctly smaller than 



!/> 



304 PROFESSOE, FRANK J. COLE 

the others, and more ventral. It has its own afferent artery quite distinct from the 
one in front. Gills normal on the right side. 

(g) Gills normal on both sides, but only five afi'erent branchials arose from the 
aorta. The last, however, on each side split almost immediately to supply the last 
two gills. 

(h) Seven gills on the left side. — Other conditions exactly as in (c). 
{i) Seven gills on the left side. — Other conditions as in (c). 

It thus appears that the usual variation is the presence of an extra gill on the left 

side, the ductus oesophago-cutaneus being still present and opening independently. 

n fe '^^^^ ^^^^ described by Howes is one such, but the one mentioned by Bateson had seven 

J"*^ j^ ^^^ pairs of gills and a ductus. Howes considers that the extra left pouch is the ductus 

^ ' oesophago-cutaneus converted into a gill, so that the new feature in the varying animal 

Jl ■ is the ductus. This, of course, is quite possible, although it does not explain the additional 

Ir^ gill on the right side, where no ductus has ever been found. We have therefore in 

Myxine the potential existence of a pair of gills behind the sixth + the ductus, 

which may conceivably represent an eighth pair. 

E. The Mucous Surfaces. 

These are described from behind forwards, and as if the gut were split up along 
the longitudinal plane. 

The abdominal intestine is a typical mid-gut, and is a straight and perfectly 
uniform tube in which regions cannot be distinguished either macro- or microscopically. 
It has a maximum width of 11 mm. in a 24-cm. Hag, and is moored to the roof of the 
abdominal cavity by a median dorsal mesentery (cp. Part II., fig. l). Over the gut 
dorsally, and slightly to the right, are the gonad and the portal vein, and also the 
so-called sympathetic nerve. The latter is the most median of the three ; then comes 
the vein, and finally the gonad, which lies at first over and then laterally to the 
vein, as shown in the above figure. The walls of the vein are very thin, and the 
vessel is seen rather by its contents than by its walls. At more or less regular 
intervals the arterial supply is seen to pass on to the gut via the mesentery. 

The mesentery is easily detachable from the mid-gut in preserved animals, and 
the gut itself, where its form is not distorted by the development of the ovarian 
portion of the gonad, and in front by the liver, is smooth, and almost spherical in 
transverse section. Its outer wall exhibits an elaborate pattern due to the ramification 
of the very extensive vascular supply of its curious lymphoid coat. This pattern 
characterises practico,lly the whole length of the abdominal gut, but it is somewhat 
simplified for about 6 mm. in front of the anus — in other words, the lymphoid tissue 
is wanting in the region of the hind-gut, and it simplifies again gradually in the 
neighbourhood of the posterior lobe of the liver, and may be lost altogether about 
10 mm. behind the opening of the bile duct. 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 305 

If an incision be now made along the line of the portal vein, and the gut pinned 
out so as to expose the mucosa, it will be seen that the latter is only very slightly 
attached to the submucous coat in preserved material (cp. Part II., fig. 1), with 
the result that it very readily comes away. It is thrown into about ten prominent 
longitudinal zigzag folds. Most of these folds are continued directly into those of 
the cloaca, but dorsally, above the anus, the cloacal folds are independent structures, 
and also ventrally there are smaller secondary folds developed between the larger 
intestinal continuations. Very few of the folds of the mid-gut, in some specimens 
none at all, pass straight from one extremity of the gut to the other. They may 
bifurcate and join up again, so as to form an elongated loop, or they may bifurcate 
without rejoining, which latter happens more commonly anteriorly than posteriorly. 
Branches may be given off which themselves bifurcate, and this often occurs in 
the region of the liver, so as to produce there a more complex pattern. 

Opposite the posterior extremity of the posterior lobe of the liver the gut begins 
to change its calibre,* and the folds to flatten and die away, so that for a short 
distance behind the opening of the bile duct the mucosa is thinner, and almost, but 
never quite, smooth. However opposite, or even slightly behind, the biliary aperture 
the folds rapidly increase in size and projection, and a short distance in front of the 
opening they again project considerably into the lumen of the gut. The aperture 
of the bile duct is situated directly in the course, and breaks the continuity, of one 
of the ventral folds. 

The zigzagging of the folds of the intestine is often emphasised by very short 
transverse folds, which project from the apex of one bend, and fit into, but never 
join, the depression of a bend in the contiguous fold. The zigzags of neighbouring 
folds therefore alternate. The pattern of the mucosa is, however, best seen by 
removing it entire, which is only too easily done in the preserved gut, and examining 
its submucous surface. We then notice that the same pattern is exhibited, only, 
of course, in the form of a cast, in the submucosa itself, and here the short transverse 
folds are indicated in a very striking manner. 

According to Maas (39), the section of the gut anterior to the abdominal intestine 
may be separated into two regions — an anterior region, or true oesophagus, extending 
from the ductus oesophago-cutaneus to the point of entrance of the gut into the body 
cavity, and a posterior indiff"erent region, or "'stomach," the hinder boundary of which is 
the opening of the bile duct. The former region has about six folds, and its lining 
epithelium is stated by Maas to be of a character transitional between the many-layered 
"ecto-" and the single-layered " endo-dermis," whilst the lining of the "stomach" 
is truly endodermal, although its sub-mucosa is not that of typical mid-gut.t 

The gut, which has narrowed down considerably by the time the opening of the 
bile duct is reached, is narrowest of all just in front of this opening, and it is here 
closely invested by the cardiac portion of the M. constrictor branchiarum et cardise. 

* In most specimens it widens considerably here (cp. p. 311). t But cp. p. 312. 



306 PROFESSOR FRANK J. COLE 

There can be no doubt that the folds of this portion of the gut are directly continuous 
with those of the abdominal intestine (and doubtless, therefore, have developed from the 
same embryonic layer), which have never entirely disappeared in the flat region behind 
the bile duct. This section, however, is only continued as far anteriorly as the opening 
of the ductus oesophago-cutaneus, from which point forwards the character of the 
mucosa quite changes, and, judging from anatomical evidence only, might have developed 
from a different embryonic layer. Two at least of the folds are continued forwards 
directly on to the posterior lining of the ductus oesophago-cutaneus, where they are pro- 
longed outwards to its external opening. The other folds of the ductus are some of 
them intrinsic, and some are continuous with folds belonging to the next anterior 
region of the gut. The ductus does not open directly into the gut, but rather into a 
small pocket-like evagination of the latter, which itself then opens into the gut at 
right angles to the ductus. 

In front of the ductus oesophago-cutaneus is the branchial gut of Maas. Its 
posterior boundary is the ductus, and its lining epithelium is quite "epidermal" in 
character, which is somewhat surprising, seeing that it cannot be stomodseal in origin, 
even if it represents morphologically a greatly elongated pharynx. Its mucosa exhibits 
more numerous, closely-set, and shallow^er folds, some of which are continuous, or fuse, 
with those of the afferent gill ducts. A disturbing factor here appears to be the 
apertures of the latter ducts, and the mucosa of the branchial gut becomes increasingly 
irregular from before backwards. This, however, cannot be entirely due to the exit 
of the gill ducts, since between the first two the mucosa presents almost tlie same 
appearance as it does in front. But behind this point the pattern is complicated by 
the folds here and there gradually approximating, joining, or being connected up by 
numerous short, shallow, transverse ridges. The result is that there is a suggestion 
of a honeycomb mucosa at this region of the gut. Here also the wall of the gut is very 
thin, and the mucosa not detachable as in the abdominal intestine. 

In front of the gill region, which must, from its development, be regarded as 
secondary hrancliial gut, the mucosa presents a very regular appearance, and there are 
about fifteen moderately prominent and perfectly straight longitudinal folds. These 
folds, however, do not always pass continuously from one end of this section of the gut 
to the other, but here and there die down, and are replaced by others. Opposite the 
posterior free extremity of the pharyngeal velum the folds tend to anastomose so as to 
form a simple honeycomb pattern, and they here terminate — none of them being 
prolonged into the mouth. 

The velum, or pharyngeal valve, according to Huxley, marks off the posterior 
boundary of the mouth. It may be described as an extensive fiat dorsal duplication of the 
mucosa, situated just behind the dental apparatus, when the latter has been withdrawn, 
and attached to the mucosa of the roof of the mouth by its mid-dorsal surface i)i much 
the same way as the gut is suspended by the mesentery. Now, this suspensory fold 
is prolonged backwards behind the region of the velum, and passes insensibly without 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 307 

any break into the median dorsal fold of the oesophageal mucosa. The velum may thus 
be, and actually is, in one sense, merely the differentiated anterior extremity of the 
median dorsal fold of the oesophagus which has developed an elaborate skeletal support. 
On the other hand, the velum may represent the original roof of the gut cut off by 
lateral evaginations (cp. fig. 11). This agrees rather with the anatomical facts, and 
with what we know of the development of this structure. 

The lateral edges of the velum are curled over sharply dorsally, and its posterior 
edge, when spread out, exhibits a short, blunt, median projection stiffened by the posterior 
transverse velar bar with its irregular posterior processes, and also on each side a 
pointed projection, the inner edge of which is supported by the termination of the 
internal lateral velar bar, and the outer or turned-over edge is strengthened by the 
external lateral velar bar. The ventral or flattened portion of the velum is sustained 
by the internal lateral velar bars and their anastomoses, whilst the dorsal doubled-over 
portions are stiffened at their edges by the external lateral velar bars. The median 
suspensory portion of the velum, which is more definitel}^ developed behind than in 
front, is supported by the supra-pharyngeal skeleton associated with the anterior 
transverse velar bar (cp. Parts I., II., and III.). 

The mucosa covering the velum is smooth. 

If the roof of the mouth be examined, we find the mucosa thrown into coarse folds 
— probably due at least partly to contraction, since with preservatives such as Perenyi's 
fluid they are less obvious. Laterally in front are the conical third and the obtuse 
fourth tentacles. Just behind the latter in the mid-dorsal line will be found a pit, and 
rising from the bottom of this pit is an elevation of the mucosa, from the apex of which 
emerges the strongly curved, slender, median dorsal tooth. Immediately behind the 
level of the posterior margin of the dental apparatus (in the retracted condition) is a 
large deep recess, the ventral wall of which is formed by the mucosa of the roof of the 
mouth, which apparently terminates here in a somewhat irregular border, as figured by 
W. K. Parker (PI. XIII., fig. 7), and the dorsal wall of which is formed by the base 
of the velum. This is the recess into which the naso-pharyngeal canal opens, and the 
latter aperture is seen if the free border above is drawn forwards. It is an elongated 
oval opening, wider in front than behind, and with an indefinite posterior border, 
owing to its being prolonged on to the base of the velum as a furrow, on which it is 
continued almost as far backwards as the anterior transverse velar bar. The mucosa, 
in fact, is doubled to form the border mentioned above, then becomes continuous with 
the mucous lining of the naso-pharyngeal canal at the posterior opening of the latter, 
is next continued on to the velum, and finally passes into the mucosa of the roof of the 
mouth and oesophagus behind the velum. On each side of the base of the velum 
dorsally and laterally is a deep, forwardly projecting pit with an almost smooth lining. 
Into the blind end of this pit a fold of mucosa projects in front. Sections, however, 
show this to be due to the root of the external lateral velar bar. The pit probably 
disappears when the dental apparatus is everted, and perhaps, therefore, only exists 



308 PROFESSOR FRANK J. COLE 

when the latter is withdrawn. It is mentioned by J. MtJLLER. I do not find any 
special histological feature associated with it. 

The lining of the naso-pharyngeal canal behind the nose is quite smooth and 
without valves. When the canal is split up and the nose examined from below, it is 
found that the median olfactory lamina is deeper and thicker, and has a more 
pronounced free margin than the others, and thus, in a sense, separates the apparently 
single nasal organ into two. The posterior portion of the third lamina from the middle 
on each side appears to have a somewhat valvular character, and would tend to direct 
water upwards among the laminae in general. The lining of the nasal tube in front 
of the nose is almost smooth, and has no definite folds ; but there is a well-defined 
transverse fold and a slight median dorsal longitudinal fold just in front of the nose 
itself, whilst dorsally and medially, just inside the external nasal opening, there is a 
blunt projection supported by a small independent cartilage (fig. 1, present part; and 
figs. 1, Parts I. and III.). 

From the Jloor of the mouth there projects into its cavity the dental apparatus, 
consisting of two halves separated by a furrow or gutter. The mucosa is reflected over 
the bases of both rows of teeth, but more particularly over that of the outer row, whilst 
the two rows of each side are separated by another and very distinct fold of mucosa, 
the tips of the outer teeth being almost on a level with, or projecting slightly beyond, 
it. The dental apparatus is obviously a derivative of the mucosa of the floor of the 
mouth, and cannot move independently of it. Therefore, to admit of any movement 
at all, the mucosa in front of and behind it must be loose and ample. When the 
apparatus is withdrawn there is an obvious transverse folding or puckering of the 
posterior mucosa (which was previously on the stretch), whilst the mucosa in front is 
quite taut. Behind this give-and-take region the mucosa of the floor of the mouth is 
thin and almost smooth. Again, the margin of the mouth is always puckered when 
the dental apparatus has been withdrawn, but when it is everted there is a pleated area 
ventrally outside the mouth, which is not fully shown in any figure I have seen. In 
the retracted condition there is a raised zone of mucosa extending straight forwards over 
and in front of the protractor tendon, and which in places is very sharply folded — in 
fact, so much so that one often has to separate the lips of the depressions with a needle 
before they are noticed. These occur nowhere else in the mouth. They seem to me 
to be due to the fact that when the pre-dental mucosa is drawn into the mouth it is 
compelled to occupy a much narrower space laterally than it does when outside 
the mouth. 

The mucosa of the floor of the mouth is practically smooth. 

F. Histology of tru Gut. 

The histology of the alimentary tract of Myxine has already been described and 
figured by K. E. Schreiner, and especially by Maas. The latter has investigated the 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 309 

structure by the new methods of solution and digestion, using caustic alkali and 
pancreatic extract. I shall therefore confine myself to a general statement of the 
anatomy of the gut, supplemented by such new details as are necessary to amplify and 
elucidate the work already accomplished. 

The gut of Myxine passes straight from mouth to anus. There are no obvious 
divisions into stomach, duodenum, and intestine. It is characterised by one very 
striking and unique feature. The mucous epithelium from the mouth to the entry of 
the gut into the abdominal cavity — that is, rather more than a third of its entire length 
— exhibits the peculiar and characteristic structure of the epidermis. The obvious 
explanation of this is that the whole of the lining of the anterior part of the gut is 
stomodseal in origin, but this we know to be not the case. The only other possible 
explanation, therefore — which is in fact no explanation at all — is that we have here a 
remarkable example of what is known as convergence. There can be no doubt that 
this anterior third of the gut represents an enormously elongated pharynx. It is 
pharyngeal in position in the embryo, but the interpolation of the club muscle during 
development detaches the gill-bearing region of the gut from the mouth, and results in 
the curiously posterior position of the gills in the adult. 

There are two types of gut structure in Myxine. These we may call the pharyngeal 
gut and the abdominal gut. The former stretches from the mouth to the entrance of 
the gut into the abdominal cavity, and the latter continues the gut to the cloaca. 

As an example of the 'pharyngeal gut or " (esophagus " we may take what Maas 
calls the branchial gut. The mucous epithelium, which is many-layered, has just the 
character of the outer skin, and possesses in abundance both the clear glassy and the 
granular slime cells characteristic of the epidermis. Maas distinguishes a stratum 
corneum and a stratum Malpighi, but, unless these terms are to acquire a new meaning, 
I do not see how they can be applied in this case. He also describes a muscularis 
mucosae which goes far into the folds ; but I find no trace whatever of this, nor, 
apparently, does Schreiner. The submucosa is dense and fibrous, and with many 
lacunar blood spaces. According to Maas, however, it consists of loose, very uniform, 
reticular cells with no lacunae. Externally there is an obvious circular musculature of 
unstriated fibres. These fibres course among the dense connective tissue of the sub- 
mucosa in its peripheral region. There are only a few isolated fibres, but they are easily 
seen, and they course right round the gut, taking no account of the folds of the mucosa. 

According to Schreiner, the mucosa has a basal membrane and exhibits no trans- 
verse folds in the " oesophagus," but these are certainly present in the branchial region. 
The superficial cells of the mucous epithelium give the slime reaction, and have a thin 
homogeneous cuticle. He finds a coiled thread in the granular slime cells, exactly as in 
the corresponding cells in the epidermis. I am unable to confirm this, in which I agree 
with Haack. There are numerous mitoses in the mucosa. , 

In the abdominal gut, it is stated by Maas that the folds of the mucosa have less 
projection, although this is not so in my preparations. He finds the ventral folds are 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART IL (NO. 3). 40 



310 PROFESSOR FRANK J. COLE 

the better developed, but I agree with Schreiner that the dorsal folds are usually the 
stronger. Ma as holds that the folds of the abdominal gut are not to be horaologised 
either anatomically or physiologically with those of the pharyngeal gut, basing the 
distinction on the behaviour of the muscularis mucosae, which I do not find anywhere in 
the gut, and on the assumption that the stratum compactum is confined to the abdominal 
gut, with which I do not agree. His assertion, therefore, that in the one case the folds 
of the mucosa do not affect the underlying tissues, but do in the other, is one which may 
well be questioned. In 8-9 cm. Hags he finds the abdominal folds only faintly defined. 

The mucous epithelium of the abdominal gut is entirely different from that of 
the pharyngeal gut. It is single-layered, and there are no traces of the glassy and 
granular mucous cells. There are, however, numerous highly granular unicellular 
gland cells, which are highly eosinophilous. The free border of the epithelium is 
striated ; and nearer the lumen end of the cells, i.e. outside the nuclear zone, there are 
many diffuse mitoses. The mucosa is pitted, and is very easily detached from the 
submucosa owing to the extensive lymph sinus between the two layers. Both Maas 
and Schreiner describe a slender basement membrane which I have not found. It is 
possible they may be referring to a membrane between the mucosa and the stratum 
compactum, and which appears to me to belong to the latter layer. 

There is no muscularis mucosae, but in its place Maas describes what he calls the 
peculiar and characteristic stratum compactum, which is highly vascular, and dispatches 
supporting processes into the adjacent lymphoid tissue. Maas has devoted consider- 
able attention to this tissue, without, however, noticing that it is strictly comparable to 
the submucosa of the pharyngeal gut. His digestion experiments establish that it is 
not one of the elastic connective tissues, and he holds that the connective-tissue frame- 
work of the abdominal gut is more specialised than that of the pharyngeal gut, since 
the embryonic formative cells have more and more receded in favour of the cell product. 
The obvious and important blood sinus which is seen in the stratum compactum 
immediately under the mucosa is usually flushed with blood, and has a definite lining 
with longish flattened nuclei. Maas w^as unable to find this lining. 

External to the stratum compactum is the lymphoid portion of the submucosa, and 
then follow the weak circular unstriated musculature (no longitudinal musculature) and 
the serosa with its attached connective tissue. 

The lymphoid submucosa is very vascular, and is continued into the folds of the 
mucosa. Its characteristic feature is the presence of packets of lymph cells associated 
with the factors of the portal vein, but not with the arteries. Such lymphoid tissue 
in the wall of the gut occurs in a very few forms {e.g., Protopterus) , and Maas regards 
it as a diffuse spleen, holding that the compact spleen of the higher forms has been 
produced first by the concentration of such a tissue, and then by its emergence from, 
the wall of the gut. Maas, who overlooked the essentially adipose nature of the 
lymphoid zone, observed by Schreiner shortly before, states that the lymphoid heaps 
only occur in the more peripheral regions, and then specially round the smaller veins. 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 311 

On the other hand, I find them in all parts of the layer, and round veins of all sizes. 
They form a kind of pulp compared by Maas with the splenic tissue of higher 
vertebrates. He states that the lymphoid cells resemble the free colourless corpuscles 
of the blood, to which they undoubtedly give rise : not, however, to the red corpuscles, 
which always lie free in the vessels and lacunae, and the early stages of which are 
never found in the connective-tissue framework of the gut. On the other hand, the 
early stages of the leucocytes occur in the framework, and the lymphoid tissue of the 
gut represents their principal source, the associated veins always containing an excess 
of leucocytes. It is, however, difficult to believe that the lymphoid tissue performs 
this function only. It is so extensive a structure that some further explanation of its 
presence must be sought. 

SoHREiNER describes mitoses both in the epithelial and glandular cells of the 
abdominal mucosa. Gland cells are never formed from epithelial cells, but are always 
derived from pre-existing gland cells. The nucleus migrates towards the lumen before 
it divides, and this explains why the mitoses are always found outside the nuclear zone. 
ScHREiNER finds the lumen end of the cell rich in fat droplets, which diminish when 
the animal has been kept in an aquarium some days without food. This part of the 
cell gives the slime reaction with appropriate stains. I agree with Schreiner that the 
thick free edging is in two parts — a proximal, binding the cells together ; and a striated 
distal, of elements corresponding to the individual cells. 

Both Maas and Schreiner state that the abdominal gut is uniform in structure 
from one end to the other. I have carefully explored the entire length of this section 
of the gut, and agree to the above statement with, however, this proviso : I have 
generally found that, either immediately or shortly following the bile duct, the lumen 
of the abdominal gut undergoes expansion — largely owing to the fact that the mucosa 
is at this point without folds and almost smooth. The entire wall of the gut, in fact, 
is here very thin. On the other hand, the above features are not associated with any 
histological peculiarity. Nevertheless, this region must have some significance. 
Further, at the hind end of the abdominal gut, the mucosa is thrown into numerous 
secondary folds, in which the suhmucosa takes no part. 

Such are the characters of the two types of gut structure found in Myxine. We may 
now consider the various stretches of the gut in more detail. 

Maas distinguishes the following regions in the gut of Myxine, apart from the 
mouth : — 

1. Branchial gut — up to the opening of the ductus cesophago-cutaneus 30 

2. True oesophagus — up to the entrance of the gut into the body cavity 3 '5 

3. Stomach— up to the opening of the bile duct . . . . 2*5 

4. Mid-gut or abdominal gut 60-5 

5. Hind-gut 3*5 



100 



* These figures are not Maas', but are calculated from data given by him. 



312 PROFESSOR FRANK J. COLE 

It will be noticed that Ma as does not regard what I have termed the pharyngeal 
gut as the cBSophagus, and in this I entirely agree with him. There is something to be 
said in favour of considering that portion of the gut extending from the ductus 
oesophago-cutaneus up to the entry of the gut into the abdominal cavity as the true 
morphological oesophagus. Maas' description of the region, however, does not always 
agree with what I have found. He describes regular folds, the interior of which is 
filled with adenoid lacunar connective tissue. I find the folds anteriorly are twice as 
high as those of the branchial gut, and higher still behind. Their submucosa is very 
dense and fibrous, and not difi"erent from that of the branchial gut. As regards the 
epithelium, he says it is layered, but with not so many layers as in the branchial gut ; 
is in two divisions corresponding to the stratum corneum and stratum Malpighi (the 
nuclei being difi'erent in these two layers) ; and that there is a striking diminution in 
the slime cells. In my preparations there are more layers in the epithelium of the 
"true oesophagus" than in the branchial gut, and Maas' statement only applies to the 
posterior end. I cannot distinguish the two layers, nor in fact any essential difference 
between the epithelium of the oesophagus and the branchial gut, in both of which the 
glassy and granular mucous cells are very abundant. Posteriorly, however, the 
granular cells are greatly reduced in number, although the glassy cells are still present 
in quantity. The superficial cells here give the mucin reaction, and the epithelium is 
more like that figured by Maas. I find no trace of the muscularis mucosae described 
by Maas, but the unstriped circular musculature is very well marked. Maas' striped 
musculature is, of course, the constrictor cardia3, and does not belong to the intrinsic 
musculature of the gut. 

I see no grounds for regarding that part of the gut which extends from the entrance 
of the gut into the body cavity up to the opening of the bile duct as a stomach. The 
bile duct does not coincide with any change in the structure of the gut, and therefore 
to use it as a boundary is entirely arbitrary. Maas gives the characters of this part 
of the gut as follows : Epithelium single-layered, with striated border and- no pits. 
Gland cells and mitoses present, also a muscularis mucosae. Adenoid submucosa 
with large vessels. Circular unstriated musculature. In my preparations the striated 
border and gland cells are not present at the anterior end of the "stomach," but 
epithelial pits do occur in this region of the gut. I find no muscularis mucosae, and 
the submucosa in every sense connects up the dense fibrous submucosa of the 
oesophagus with the stratum compactum of the abdominal gut. 

ScHREiNER divides the. gut of Myxine into a mouth, oesophagus [pharyngeal gut], 
intermediate region (no stomach), true gut, and ectodermal anus and cloaca. In the 
mouth he finds the epithelium many-layered, the superficial layer giving the slime 
reaction, and the basal cells being small and polygonal. The slime cells, the life of 
which terminates with the discharge of the slime, are exactly similar to those of the 
skin, and develop in the same way. There are no granular cells. The boundary 
between the mouth and "oesophagus" is the naso-pharyngeal opening. In the "inter- 



ON THE GENERAL MORPHOLOOY OF THE MYXINOID FISHES. 313 

mediate region," which may be entirely closed by the constrictor muscle [constrictor 
cardiie], the granular cells become rare and finally disappear, the slime cells are reduced 
in number, and the superficial cells again give the slime reaction as in the mouth. 

Haack, who appears to be unaware of Maas' paper on the gut of Myxine, states 
correctly that the multicellular oral gland of the lamprey is not present in Myxine. 
He criticises Schreiner's paper as regards some of the smaller detail, and finds no 
transverse folds in the pharyngeal gut, as I do. He regards the gut up to the 
beginning of the gills as corresponding to the pharynx of Petromyzon. My prepara- 
tions agree with his in one respect — that the granular cells do not contain a continuous 
thread, as described by Schreiner. They do not, however, always contain the thread 
in the outer skin. He compares the granular cells with the gland cells in the 
oesophagus of the Ammocoete. 

The nasal tube is surrounded more or less by an extensive lymphatic sinus situated 
between the nasal rings and the tube itself. The epithelium is many-layered and rests 
on a distinct basement membrane. Under that there is a dense fibrous vascular 
connective tissue which diminishes posteriorly. The superficial cells of the epithelium 
give the mucin reaction. The glassy mucous cells are very numerous and larger than 
those in the skin, but the granular cells are scarce and quite small. 

There are no mucous cells in the olfactory laminae of the nose, although they are 
still present in the ventral non-olfactory portion of the tube, the epithelium of which 
is so thin that the mucous cells not only occupy its entire height, but project beyond 
it into the lumen of the nasal chamber. 

In the naso-pliaryngeal tube the epithelium is very thin but many-layered, and 
there is a definite basement membrane. The glassy mucous cells are very numerous, 
but only a few of the superficial cells give the mucin reaction. The discharging glassy 
cells project markedly into the lumen of the tube, and often at the other end rest on 
the basement membrane. There are numerous granular cells in the posterior section of 
the tube, especially in the region of its opening. Next the epithelium is some very 
loose vascular connective tissue, which posteriorly becomes fibrous, and then follows 
the lymph sinus. 

The epithelium of the mouth near the opening is many-layered, and rests on a dense 
fibrous connective tissue. There is no difi"erentiated free border. The three or four 
superficial layers of cells give the mucin reaction, and their nuclei are pushed to one 
end of the cell. The glassy mucous cells are numerous and large, but there are no 
granular cells, nor any multicellular glands. There are no mucous cells in the 
immediate neighbourhood of the teeth, and in fact the number of these cells is reduced 
in this region, the lateral epithelium further being here very thin, but containing 
raucous cells. 

At about the region where the naso-pharyngeal tube opens into the mouth, i.e. 
opposite the anterior extremity of the notochcord (cp. fig. 1), large numbers of the 
granular mucous cells appear in the epithelium, similar to those in the skin, but 



314 PROFESSOR FRANK J. COLE 

smaller. They also occur in the hinder extremity of the naso-pharyngeal duct. At 
the same time the glassy mucous cells increase in number, and the superficial cells 
giving the mucin reaction so characteristic of the anterior section of the mouth 
almost disappear, to be replaced by an epithelial mosaic. Further back, however, 
they become more numerous again, but are never as well developed as in the mouth. 
There is no doubt that we have here a distinct change in the character of the 
epithelium, which corresponds to the boundary between the stomodseum and the 
mesenteron. 

The submucosa is a very dense fibrous connective tissue immediately under the 
epithelium, but becomes looser further away from it. The gut behind the velum 
resembles the velar gut. 

Posterior to the velum the gut gradually narrows down to rather more than half 
the width at the velum, and the lumen becomes correspondingly contracted and 
flattened dorso-ventrally. It then gradually increases in calibre, the lumen becoming 
larger and oval in transverse section. These changes, however, involve no corresponding 
histological change. 

There is no real distinction between the pre-branchial and the branchial gut, but 
the branchial submucosa is more vascular and not so dense, and there are rather more 
unstriped circular muscle fibres, although they are still very sparse. Again, near the 
hind end of the branchial gut I have observed a slight tendency for the unstriated 
musculature to be directed into \he, folds. Both the glassy and, to a lesser extent the 
granular, mucous cells, are still present in great abundance, but the unstriped 
musculature embedded in the submucosa almost disappears posteriorly, and external to 
the submucosa for a short distance we have a new layer, a wide zone of fatty tissue, 
encircling which is a layer of unstriated muscle. Apart from the epithelium a section 
of the gut here is not unlike one of the abdominal intestine. 

The branchial gut increases gradually in calibre from before backwards, and the 
lumen is largest a short distance in front of the ductus oesophago-cutaneus. It then 
narrows down markedly, increases again in the immediate neighbourhood of the ductus, 
whilst behind the latter, and where it is surrounded by the constrictor cardise, it 
abruptly diminishes, and is smaller and more definite in shape {i.e, circular in transverse 
section) than in any other region of the gut. Behind the constrictor it widens again 
gradually until the region of the bile duct is reached, and then it rapidly expands into 
the full size of the abdominal gut. 

The transition from the branchial into the abdominal gut may be best studied in a 
series of longitudinal sections. We then note the following points : — 

Epithelium. — No change is noticeable until about the centre of the constrictor 
cardise, where there is a marked diminution in the granular mucous cells. The glassy 
cells also decrease in number, but not to the same extent. Both kinds of cells, however, 
may occur right up to the boundary between the two types of gut — i.e. between the 
many-layered and single-layered epithelium. With the diminution of the mucous cells 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 315 

the superficial cells of the epithelium again assume the mucin-secreting character as in 
the mouth. The epithelium gradually decreases in height from 72 m up to the 
boundary, where it is only 40-48 m high. The transition is quite sudden. It does not 
occur at the same level in all the folds — otherwise a straight line would accurately 
divide the two types of epithelium. The single-layered epithelium of the abdominal 
gut is at first without the striated border, nor does it possess any gland cells. It very 
gradually increases in height, and acquires first the striated border and then its charac- 
teristic unicellular glands. 

Submucosa. — Dense and fibrous in the branchial gut, but immediately in front of 
the constrictor cardige it is largely replaced by a thick layer of fatty areloar tissue. In 
the region of the constrictor, however, the submucosa is again, and even more, dense 
and fibrous, and becomes increasingly so as it passes backwards. Behind the constrictor 
{i.e. at the epithelial change) it gradually diminishes, and passes without a break into 
the stratum compactum of the abdominal gut. There is no doubt that the latter layer 
is only another form of the same tissue as the submucosa of the pharyngeal gut, 
although its staining reactions are slightly different. The curious submucous blood 
sinus associated with the stratum compactum may be traced as far forwards as the 
epithelial change. 

Musculature. — ^The unstriated musculature gradually increases as we trace the 
branchial gut backwards, and it is a very conspicuous layer in the fatty region im- 
mediately in front of the constrictor cardiae. It diminishes somewhat at the anterior 
end of the constrictor, but increases again as it passes backwards, decreasing once more 
at the epithelial change, finally becoming directly continuous with the weak unstriated 
musculature of the abdominal gut. The new layer, therefore, in the latter gut, which 
is not represented in the branchial gut, is the thick zone of fatty lymphatic tissue 
between the stratum compactum and the unstriated musculature. 

The entrance of the bile duct marks no change in the character of the gut, and 
hence 1 am unable to agree with Maas that there is a representative of the stomach 
in Myxine. The epithelium, stratum compactum, and the lymphoid tissue are the same 
both in front of and behind the aperture of the bile duct, although the typical abdominal 
gut extends only a very short distance in front of this point. 

The passage of the mid- or abdominal gut into the hind-gut or cloaca is also best 
observed in serial longitudinal sections. We thus find : — 

Epithelium. — Insensibly diminishes from 120 m, losing its gland cells and its 
striated border, until at the boundary it is only 32-56 m high. Maas figures the 
striated border and the gland cells right up to the change, a condition not found in 
any of my preparations. The single-layered epithelium of the abdominal gut meets 
the many-layered epithelium of the cloaca at a bevel joint, dissimilar to the junction 
of the branchial and abdominal guts, and in such a way as to suggest that the abdominal 
epithelium represents the gradually heightened superficial layer of cells of the cloacal 
epithelium. The latter increases slowly in height up to 80 m, and for some distance 



316 PROFESSOR FRANK J. COLE 

is devoid of the glassy and granular mucous cells. Near the cloacal opening, however, 
and in the dorsal chamber of the cloaca,* the two latter become extremely numerous — 
in fact much more so than they are in the neighbouring skin. Posterior to the anus, 
the lining of the entire cloaca, and even the urinary papilla, contain both kinds of 
raucous cells. 

Suhmueosa. — The stratum compactum behaves as at the anterior boundary, and 
passes gradually into the dense and more deeply staining fibrous submucosa of the 
hind-gut. In the 19 -cm. Hag the submucosa of the posterior extremity of the mid- 
gut is almost replaced by large blood spaces. The characteristic blood sinus in the 
stratum compactum immediately underneath the mucosa can be traced no further back 
than the change in the epithelium. The lymphoid cells are not present at the extreme 
end of the submucosa of the mid-gut, and the adipose areolar section of the submucosa 
dies away somewhat in front of the epithelial boundary. 

Musculature. — In the sections of the 25-cm. Hag a new tissue appears in the hind- 
gut. The unstriated musculature of the mid-gut disappears, and is replaced by a 
primitive tissue resembling an extremely simple form of cartilage, unlike, however, any 
definitive cartilage of Myxine. It encircles the whole of the hind-gut, exclusive of the 
posterior extension of the body cavity and of the segmental ducts, and recalls the thin 
sheet of cartilage described by Ayers and Jackson in the wall of the cloaca of 
Bdellostoma (cp. my Part I., p. 786). In the sections of the 19-cm. Hag this tissue is 
not present, and its place is occupied by a well-developed and undoubted unstriated 
circular musculature, but there are indications posteriorly of its replacement by the 
connective tissue. 

Maas failed to find any glassy or granular cells in the cloacal epithelium, although 
he says they reappear in the " anal " slit itself He also describes a muscularis 
mucosae, a statement I am unable to confirm. Schreiner, like myself, finds both kinds 
of mucous cell in the cloaca. 

The height of the mucous epithelium in the different regions of the gut may now 
be given. The measurements are in /x. The nasal tube at the opening is 48 ; it 
rises to 100 and sinks to 60 at the posterior end. In the naso-pharyngeal duct the 
epithelium is only 32 at the anterior end, but it gradually deepens as it approaches its 
oral aperture, where it is about 80. In the mouth, up to the posterior extremity 
of the velum, the height varies from 80 to 140, but is more frequently about 100, the 
lower figure, however, being more usual behind. In the pre-branchial gut the epithehum 
is very constant, and measures almost invariably 100. In the branchial gut it drops from 
100 to 80. It rises in the region of the constrictor cardise to 100 and 120, but falls 
to 80 on the first entry of the gut into the abdominal coelome. In the abdominal gut 
the figures vary from 102 to IGO, the commonest measurement being 140, whilst at the 
posterior end it is 120. 

* i.n. ill the region of Burnk's "anal slime gland." 



OK THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 317 

G. The Liver and Biliary Apparatus. 

1. Superficial Anatomy (Figs. 4 and 13). 

The liver consists of two lobes — a large posterior and a smaller anterior. When the 
body cavity is opened by a median ventral incision the anterior lobe is observed to 
overlap the posterior, the latter on the left passing forwards under the anterior lobe. 
On the right the gall bladder projects from a notch between, and dorsal to, the two lobes 
of the liver, being partially overlapped by both, the right posterior extremity of the 
one and the anterior extremity of the other being bevelled off to receive it. The 
anterior border of the anterior lobe is just at the level of the two external branchial 
openings. 

The sub-intestinal vein, formed by factors from the ventral wall of the anterior 
abdominal gut, suddenly emerges from the gut wall opposite, or in front of, the posterior 
end of the posterior lobe of the liver, and passes downwards along the hinder vertical 
border of the hepatic ligament to reach the apex of the posterior lobe. From here it 
passes forwards towards the sinus venosus along the entire length of the ventral surface 
of the posterior lobe. A smaller branch, however, may be traced along the dorsal 
surface. These conditions, or rather their superficial relations, vary in different 
individuals. 

The mesentery, which suspends the gut, passes from the ventral wall of the gut on to 
the posterior lobe of the liver as the hepatic ligament. It supports the posterior lobe 
along its whole length, the only interruption being in the region of the gall bladder. 

The depression associated with the point of exit and entry of the posterior hepatic 
duct and vessels is situated in the direct plane of the hepatic ligament. Hence the 
two factors of the ligament, in passing straight on to the liver at this place, simply 
diverge to allow the passage of the vessels. In front and on the right, the posterior 
lobe of the liver is bevelled off dorsally to receive the gall bladder, whilst its anterior 
extremity is overlapped ventrally by the anterior lobe. The two lobes are absolutely 
distinct, each having its own independent serosa.* 

In the region of the gall bladder the mesentery passes loosely from the gut on to 
the right side of the gall bladder. This admits the passage of the various ducts and 
vessels associated with the gall bladder and the two lobes of the liver. From the left 
side of the gall bladder the mesentery is reflected on to the dorsal surface of that portion 
of the posterior lobe of the liver in front of the depression mentioned above. 

Unlike the posterior lobe, the anterior lobe of the liver has no definite ligament, but 
its posterior edge usually more or less adheres to the posterior lobe. The mesentery 
passes on to it from the anterior left surface of the gall bladder, and reaches it immedi- 

* This statement must be qualified. In one series of sections I found a slight but quite unmistakable fusion of 
the glandular tissue of the mesial edge of the posterior lobe and the dorsal surface of the anterior lobe in the region 
of the overlap a little to the left of the middle line of the body. Further, in one dissection the two lobes of the liver 
were completely fused into one body. 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 3). 41 



318 PROFESSOR FRANK J. COLE 

ately behind the region where its vessels enter and leave. From here it extends back- 
wards, forming a separate bag in which the anterior lobe lies, and its only attachment is 
in front where the mesentery passes from almost the entire anterior edge of the lobe on 
to the pericardio-peritoneal partition, itself in its free portion consisting of a double 
sheet, i.e. a peritoneal and a pericardial factor. Elsewhere it is the latter factor which 
excludes the liver from the pericardial cavity. 

On the side of the intestine and the body wall the posterior lobe of the liver bears 
a prominent anterior obliquely curved bevel for the reception of the gall bladder. 

When the lobes of the liver are turned over to the left and pinned down so as to 
expose the whole of the gall bladder, the latter is seen to be a relatively very large sac. 
It is dorsal to the liver, and entirely on the right side of the body. Two of its ducts 
are seen at once {a.h.d. and^.A.,cZ.). They are the two hepatic ducts — really hepato-cystic 
ducts — arising one from each lobe of the liver {a.l.l. and^.^.Z.), from its dorsal surface 
in the middle line, respectively a little in front and a little behind the gall bladder 
itself (^.6.). Between these two the bile duct (h.d.) leaves the ventro-mesial surface of 
the gall bladder, and, passing somewhat forwards, soon enters the gut a little to the left 
of the raid-ventral line and a short distance behind the heart. If the intestine be split 
up, the bile duct is observed to open into it by a conspicuous aperture at the apex by a 
large backwardly directed papilla. As the bile duct approaches the intestine its wall 
becomes thickened — due to an aggregation under the serosa of the follicles of the 
supposed " pancreas," the structure of which is described elsewhere. 

On opening the gall bladder, and delicately brushing its internal surface clean, we 
observe a conspicuous depression, with an aperture at the bottom, on the surface facing 
the intestine. This is the opening of the bile duct. A short distance in front, but to 
one side of it, is the smaller and less obvious aperture of the anterior hepatic duct. A 
slight distance behind and also to one, and the opposite, side (c^^ fig. 4) is another 
similar opening — that of the posterior hepatic duct. There are thus three openings 
into the gall bladder, and the hepatic ducts have no direct connection whatever with 
the bile duct. In one specimen I found two anterior hepatic ducts, each having a 
separate opening into the gall bladder, there being therefore in this case four cystic 
openings. The additional duct, which was very small, was formed by the fusion of two 
thread-like tubes from the anterior lobe of the liver, but on nearing the gall bladder it 
became enlarged to the normal width and opened by a small aperture close to the 
larger one of the normal duct. 

The anterior hepatic duct (a.h.d.) courses almost straight backwards, until it 
reaches the gall bladder, where it bends abruptly on itself in a U and opens almost at 
once into the cyst. The posterior hepatic duct [p.h.d.) passes almost straight for- 
wards, but on reaching the gall bladder it bends over to the opposite side [cp. fig. 4), 
almost at a right angle, to reach its aperture. This difference between the courses of 
the two hepatic ducts I find also in the sections. As shown in fig. 4, each hepatic duct 
is accompanied by a branch of the common portal vein (c.jj.v.) and of the coeliac 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 319 

artery (coe.a.), the latter especially being closely attached to the duct, and somewhat 
difficult to separate out. All three respectively enter and emerge for the liver at the 
same place. The branch of the coeliac artery, dividing into the cystic and hepatic 
arteries, closely accompanies the bile duct — in this case anteriorly. The main stem of 
the artery courses along the intestine, and may pass over or under the root of the bile 
duct. In another, and well-injected, specimen the coeliac artery behaved as follows : 
it arose as a pcm-ed artery, which, after giving oif on each side various branches, e.g., 
to the pronephros, fused ventrally so as to form a complete ring around the gut in 
front of the bile duct. From this ring another and smaller circle was given off below 
the gut which surrounded the bile duct. From the latter arose the branches to the 
"pancreatic" follicles, the bile duct and gall bladder, and the anterior lobe of the liver. 
Posteriorly the lesser ring gave off a large branch to the mid-ventral surface of the gut, 
and the branch to the posterior lobe of the liver. 

In a Bdellostoma cirrhata, Sch., dissected by Dr Dakin, the anatomy of the gall 
bladder and cystic ducts was found to be essentially the same as in Myxine. 

J. Muller's figure of the gall bladder and ducts may conceivably represent one of 
the variations which are so common in Myxine, but it is certainly quite inaccurate as a 
representation of the normal condition. He figures the two hepatic ducts opening into 
the base of the bile duct. His very brief reference to this region in the text takes no 
account of the openings of the ducts. The fact that the hepatic ducts may open direct 
into the gall bladder is, of course, an unusual but not a unique phenomenon. It was, 
1 believe, first described in the ox by Verheyen in 1710, and occurs as a rare variation 
even in man. I am not aware, however, of any case so clear and interesting as that 
of Myxine. 

WiEDERSHEiM, in the first edition of his Lehrhuch (Th. ii, p. 590), publishes an 
original figure of the biliary apparatus of the Myxinoids, which agrees practically with 
Muller's figure, and to which therefore the comment above also applies. His remarks 
are, however, more definite. He says: "Aus jedem Leberlappen tritt ein Ductus 
hepaticus hervor und diese confluiren zu einem Ductus choledochus, in dessen Riick- 
wartsverlangerung die Vesica fellea (resp. der Ductus cystieus) gelegen ist." 

Even Maas, in his work on the " pancreas" of the Myxinoids, agrees practically with 
MuLLER and Wiedersheim. In Myxine he says (p. 3) : " Die Beziehungen der drei 
Gauge sind so innige, dass zwischen vorderem Lebergang und der Austrittsstelle des 
Gallenganges nur ein verschwindendes Stlick Gallenblasenwandung liegt, und dass fiir 
den hinteren Lebergang der Gallengang einfach die Fortsetzung bildet." In Bdellostoma 
he is more definite (p. 571): "Dieser letztere [bile duct] setzt sich, durch ein en sehr 
kurzen Ductus cystieus mit der Gallenblase in Communication stehend, aus der 
Verbindung von rechtem und linkem Lebergang zusammen und miindet wie bei 
Myxine genau ventral median in den Darm ein." I have repeated my dissections of the 
biliary apparatus of Myxine a number of times, and find on the whole a remarkable 
agreement with the conditions already described. The only variation of importance was 



320 PROFESSOR FRANK J. COLE 

the occurrence of the extra anterior hepatic duct mentioned above. If the gall bladder 
is split open, and its internal surface carefully examined, there is no difficulty in 
finding the three perfectly distinct apertures as I have described them. 

2. General Anatomy of the Liver. 

(A.) First, as regards the blood-vessels and hepatic ducts. The sub -intestinal vein, 
on reaching the posterior extremity of the posterior lobe of the liver, behaved in one 
case as follows : it split into two large vessels — one passing on to the ventral convex 
surface of the lobe, and the other on to the dorsal concave surface. The latter itself 
split into two almost equally sized vessels, and both, after repeated branching, were 
finally dissolved among the liver tubules. This proves that the sub-intestinal vein 
acts not only as a collector but as a distributor, i.e. that it brings blood to the liver 
just as the portal vein does. 

On the other hand, the vessel on the convex ventral surface, which represents the 
main trunk of the sub-intestinal vein, remained fairly constant in size in spite of 
frequent interchange of vessels with the liver. Large veins were undoubtedly received, 
showing that the sub-intestinal collects blood returning from the liver, and is therefore 
the 'posterior hepatic vein. Its course is always more or less superficial. 

No factors of the hepatic duct are associated with the sub- intestinal vein or its 
constituents. 

Anteriorly the sub-intestinal comes more to the surface of the lobe, and is plainly 
visible in a surface view. It leaves the posterior lobe at its extreme anterior point as 
its sole or principal hepatic vein and opens into the sinus venosus. There may, however, 
in addition, be another efferent vessel associated with the posterior lobe, opening 
separately into the sinus venosus and corresponding to the similar vessel [anterior 
hepatic vein) draining the anterior lobe of the liver. It may be called the accessory 
posterior hepatic vein, and, like the sub-intestinal, it receives blood from the irregular 
spaces everywhere present between the liver tubules, and courses along the dorsal 
surface of the lobe. I thought at first that this vessel might be the direct continuation 
of one of the dorsal branches of the sub-intestinal described above, just as the sub- 
intestinal itself is continued forwards as the principal hepatic vein of the posterior lobe, 
but the specimens I investigated with respect to this point gave no support to such 
a supposition. The accessory posterior hepatic vein, therefore, is a true hepatic vein, 
and is actually constituted within the substance of the posterior lobe of the hver. 

From behind forwards the above vessels open into the sinus venosus in the following 
order : (1) accessory posterior hepatic vein ; (2 and 3) posterior hepatic or sub-intestinal 
vein and anterior hepatic vein almost simultaneously. In front of this region the 
sinus venosus in my large series of sections received seven independent smallish 
accessory anterior hepatic veins from the anterior lobe of the liver, which extended 
forwards for some little distance in front of the place where the main anterior hepatic 
vein left it. 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 321 

There is a distinct diflfereiice between the larger branches of the portal vein and 
those of the sub-intestinal vein. In the former case they have an obvious connective- 
tissue sheath, and are surrounded by factors of the hepatic duct, whilst in the latter 
case they are irregular spaces between the liver tubules, apparently unlined,* and 
usually contain a lesser quantity of blood. Nevertheless, the two series of spaces 
communicate freely within the liver. A forward injection of the portal vein at once 
passes from the liver substance backwards along the sub-intestinal vein. 

The portal vein, on entering the liver, splits into a number of large vessels of 
considerable blood capacity compared with that of the vein itself. The pressure of 
blood in the liver, therefore, must be low. 

As that division of the portal vein which enters the posterior lobe of the liver is 
coursing backwards over the surface of the liver preparatory to plunging into it, there 
is given off first a large vessel which courses forwards in the interior of the anterior 
section of the lobe, and which preserves its identity almost up to the anterior extremity 
of the latter. The portal vein then detaches another large vessel, which supplies the 
more central section of the gland. Finally, it enters the lobe at the point where the 
hepatic duct leaves it, coursing at first alongside and external to the associated larger 
factors of the hepatic duct near the surface of the lobe, and giving off large vessels to 
the interior. It then penetrates to about the centre of the lobe, still accompanying the 
larger factors of the hepatic duct. Finally, as the posterior extremity of the lobe is 
approached, and the hepatic duct is represented by a large number of very small 
tubules, the portal vein splits first into two and then into four, and so on until its 
identity is no longer maintained. 

With regard to the anterior division of the portal vein to the anterior lobe of the 
liver, the vessel, on entering the lobe, gives off two large branches which pass back- 
wards, and supply that part of the gland behind the point of exit of the hepatic duct. 
The remainder passes forwards, accompanied by the factors of the hepatic duct, soon 
splits into two and then into four, and thereafter behaves in the same way as in the 
case of the posterior division. 

Near the point of exit from the liver we encounter the usual portal canals, con- 
sisting of one or more branches of the portal vein, hepatic artery, and hepatic duct, 
enclosed in the connective-tissue sheath of Francis Glisson. This association, however, 
has by no means the definite significance and regularity of the higher animal. The 
artery is distinguished by a well-defined lining and its conspicuous closely set nuclei 
projecting into the cavity of the vessel. 

The, posterior hepatic duct, after leaving its lobe of the liver, receives a large duct 
from the anterior section of the lobe. If this itself be now traced forwards, it almost 
at once, and before becoming actually buried among the liver tubules, splits into a 
very large number of small ducts which are arranged round the portal vein and its 

The boundaries exhibit a few flattened nuclei at intervals which belong to the inconspicuous lining of the 
vascular space and not to the basement membrane of the tubules. 



322 PROFESSOR FRANK J. COLE 

branches. These ducts are readily identified by their more deeply staining shallow 
epithelium and closely crowded nuclei.* The contrasted liver tubules, of course, open 
into them. 

The main or posterior section of the posterior hepatic duct, on being followed 
backwards from its point of exit from the liver, is found to behave in precisely the 
same way as the smaller anterior section. Whenever the portal vein gives off a 
large branch, the hepatic duct detaches a corresponding branch, and this immediately 
breaks up into a large number of small tubules which at once surround the vein. There 
is, however, one difference between the two sections of the posterior hepatic duct. The 
posterior one preserves its identity for some distance within the liver — in fact, as long 
as the portal vein is represented by a conspicuous vessel. But when the latter splits 
into two, the hepatic duct is thereupon dissolved into numerous small tubules. 

In places the biliary tubules become enlarged, and here contain a substance which 
looks very like broken-down blood corpuscles. It is possible this may be a pathological 
manifestation. 

The anterior hepatic duct leaves the anterior lobe of the liver as two large tubes 
which, however, immediately join. Each is formed by the junction of two ducts — in 
one case of two relatively small ducts. These then enter the liver in company with 
the portal vein. Since the greater part of the lobe lies anterior to the exit of the 
hepatic duct there is not the same occasion, as in the case of the posterior lobe, for the 
existence of anterior and posterior sections of the duct, and, as a matter of fact, the two 
tubes mentioned above are concerned mostly with that part of the lobe anterior to the 
point of exit. A few smaller tubes pass backwards to drain that part of the lobe 
posterior to the exit place. The factors of the anterior hepatic duct behave, and are 
related to the anterior branch of the common portal vein, exactly as in the case of the 
posterior lobe already described. 

The liver is extremely vascular. It contains large, sometimes very large, irregular 
blood sinuses, only partially filled with blood, which at first sight appear to have no 
definite walls. As above stated, other blood spaces have an obvious connective-tissue 
wall. The former are associated with the sub-intestinal and hepatic veins, the latter 
with the portal vein. In serial sections I have been able to trace a direct continuity 
between these two types of spaces. 

In a well-injected liver each tubule is seen to be completely surrounded by a 
vascular space — in fact so much so that the substance of the liver appears to include 
more blood tissue than liver tubules. None the less, I have found no clear case of a 
vessel entering a tubule — they remain always outside them. There are no inter- 
cellular or intra-tubular vessels. 

Injection from the heart {i.e., via the hepatic arteries) or from the portal vein gives 
the same .result. A coarse vascular network is disclosed which penetrates everywhere 
between the liver tubules, and constitutes not a capillary system but a reticular sinus. 

* In one case I found only three cells in transverse section. 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 323 

This sinus is usually injected if the needle of the syringe is simply plunged into the 
liver at random. The liver admits of considerable vascular distension. The tubules 
may be so close together as to obliterate, or almost obliterate, the intervening blood 
spaces, or the latter, if congested, force the tubules apart and form relatively large 
channels. 

(B.) Second, as regards the histology of the liver tubules. The Myxinoid liver is a 
simple tubular or racemose gland, in which the tubules end blindly and form no net- 
works. Microscopically there is only one kind of tubule. I have searched in vain in 
my serial sections for any glandular tissue which might constitute a histological pancreas. 
The liver is enclosed in the usual serous and fibrous coats, the latter with long nuclei ; 
but no trabeculse are dispatched internally to ramify among the liver tubules. In fact, 
the liver has only an extremely slight connective-tissue framework. Some of the large 
superficial blood-vessels, however, are invested by fibrous tissue continuous with that of 
the external fibrous coat. 

The liver tubules are circular or oval in transverse section, and possess a very thin 
non-cellular basement membrane. The central cavity, varying with the state of activity 
of the tubule, may be large and irregular, small and round, or apparently wanting 
altogether. Under the hig-hest magnification the free border has a well-defined margin 
outlined by chromophilous granules like exceedingly minute nuclei. The diameter of 
the tubules varies from 32 to 78 m — average 56 fi ; height of the cells from 8 to 28 m 
— average 17 m;* diameter of the cavity from 4 to 44 m (Holm gives 3 to 8 m) ; 
the number of cells seen in a transverse section of a tubule is from 4 to 15, average 
number 8 to 9 (Braus says 4 to 6 and Holm 5 to 10). 

The cytoplasm of the liver cells may be very vacuolated. Large intracellular cavities 
may occur which themselves open into the larger lumina of the tubules. Intercellular 
canaliculi or bile capillaries may be very numerous and striking, so that in transverse 
section a star-shaped cavity is seen. The cells of the whole liver may be thus vacuo- 
lated, but here and there small aggregations of tubules are found with solid cytoplasm 
and slightly larger nuclei. These are in sharp contrast with the surrounding vacuo- 
lated cells, owing to their naturally taking on a deeper stain. 

Small patches of adipose tissue are to be seen scattered among the liver tubules. 

The nuclei of the liver cells are spherical, with well-marked reticular chromatin, and 
are obvious diff"erentially staining nucleolus. They measure from 5 to 10 /x, but the 
larger sizes are the more frequent. Occasional mitoses, of the type described by the 
ScHREiNERS in other Myxine tissues, may be observed. 

In thick sections we notice that the superficial tubules are looser, have larger cavities 
and more vacuolated cytoplasm, and are not so packed together as are the more central 
tubules. 

The structure of the liver of Myxine has been investigated by Braus, who used 

* According to Braus, 30 fi. 



324 PROFESSOR FRANK J. COLE 

material down to 7 cm. in length. He describes cytozonal networks formed by the 
bile channel splitting, surrounding little islands of several cells, and then joining up 
again. I also find these. Braus confirms Retzius that the innumerable small blind 
side capillaries given oS" from the central lumen of the tubule, which penetrate between 
the liver cells, do not reach the periphery of the tubule. This is generally, but not 
universally, true. He also states that, besides the nucleus, the body of the cells contains 
large "nebenkerne" resembling corresponding structures in pancreatic tissue. He 
believes these to be the remains of the achromatic spindle of recent mitoses, and thus 
archoplasmic in character. M^hilst agreeing that the liver of Myxine is a tubular gland, 
he describes certain deviations from the purely tubular structure which suggest the 
origin of networks of bile capillaries. 

Holm states that the liver of Myxine is a typical tubular gland. He describes a 
strong layer of unstriated muscle fibres round the larger blood-vessels which I have not 
found, and asserts that the blind intercellular bile capillaries reach the periphery of the 
tubule. He has injected the liver from the portal vein. 

3. Bio-chemistry of the Liver. 

I have not found it possible to overcome satisfactorily the practical ditiiculties in 
the way of testing the secretion of the so-called " pancreas " of Myxine. The gland 
itself is so very small that almost hundreds of dissections would be necessary, and its 
situation is such that one could never guarantee any extract made from them would be 
free from the secretion of the liver. In any case, however, this does not matter. As 
far as my material is concerned, the " pancreas " can have no eff'ect on proteid digestion 
in the animal. 

An eff"ective pancreas, therefore, being absent, the question is raised as to the 
function of the liver. Two problems had to be solved : ( 1 ) Was the liver a true liver- 
could it be compared with the liver of higher vertebrates? (2) Was it more than a liver 
— i.e. was it also a digestive gland ? To the former point an answer in the affirmative 
can be given. I had hoped to have included here an account of the digestion experi- 
ments which have been carried out on the liver and gut of Myxine. These, however, 
are not yet complete, and cannot now be completed until next year. The Hags 
migrate into deeper water in the winter, and at the time of writing (September 23) 
the females have already left, only one out of the 150 examined being a female. 

The chemical and physical properties of the contents of the gall bladder were 
investigated in several freshly killed Myxine. I ought to add that these specimens had 
been kept in the tanks at Cullercoats for some time, and it is well known that Myxine 
do nob maintain their health in caj^tivity, nor will they feed. The following results 
might lience have to be slightly modified if bile taken from animals fresh from the sea 
were examined. Exactly similar experiments were carried out with a specimen of bile 
taken from a freshly killed ox, the two series of observations being performed side 
by side. 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 325 

A careful comparison of the results obtained shows that Myxine bile, although 
obviously differing in percentage composition, contains all the essential constituents of 
ox bile, and hence the liver of Myxine is a true vertebrate liver, whatever else it may be. 

Both fluids have a greenish brown colour, but whilst ox bile is a clear viscous fluid, 
Myxine bile is almost opaque. The latter, however, is of a much more watery con- 
sistency, due to the presence of a smaller percentage of mucin. Myxine bile has the 
characteristic bitter taste of ox bile, but the latter is more aromatic. Further, ox bile 
is slightly alkaline to litmus paper, but Myxine bile is neutral. 

The presence of mucin and nucleo-protein may be tested for by adding a few drops 
of acetic acid to a small quantity of bile, when a precipitate insoluble in excess of acid 
is thrown down. Ox bile gave a bulky mucinoid precipitate with this test, but with 
Myxine bile the precipitate was only sufficient to render the fluid turbid. With 
methylated spirit both fluids gave a bulky precipitate of mucinoid nucleo-albumin 
coloured by pigment. 

The pigment tests were not, on the whole, quite so pronounced as with ox bile, but 
were sufficiently good to leave no doubt as to the presence of bile pigments in Myxine 
gall. 

On shaking up a small quantity of ether with the two separate samples of bile, no 
pigment was extracted by the ether, but, on the addition of a few drops of concentrated 
hydrochloric acid, nucleo-protein was in both cases precipitated and pigment extracted 
by the ether. Ox bile gave much the better result, both in the amount of the precipitate 
and of the pigment extracted. Pigment was further tested for by Gmelin's ring test. 
This is carried out by saturating a portion of filter paper with bile, and placing a drop 
of fuming nitric acid in the centre of the saturated area. After a short time a number 
of differently coloured concentric rings appear at the margin of the nitric acid drop, and 
slowly spread. The order of the colours is (1) yellow, appearing at the junction of bile 
and acid and due to the pigment choletelin ; (2) red ; (3) violet, passing into blue, due 
to bilicyanin ; and (4) green, due to biliverdin. Owing to the limited quantity of 
Myxine bile available, only a very small area of the filter paper could be saturated, but 
rings of yellow, red, and blue could be faintly seen. Ox bile, again, gave the more 
striking result. 

As regards bile salts Pettenkofer's test gave an excellent result with Myxine bile. 
The test consists in adding a drop of 1 per cent, solution of cane sugar to a few drops 
of bile, and thoroughly shaking. Concentrated sulphuric acid is then carefully poured 
down the side of the tube, and at the junction of the two liquids a purple ring slowly 
develops. On shaking, the contents of the tube are coloured a deep purple. By 
using a solution of cane sugar and sulphuric acid only, by way of control, no purple 
colour was produced, but only the characteristic brown colour due to the charring of 
the sugar by the sulphuric acid. Hay's test depends on the presence of bile salts 
reducing the surface tension of that liquid. When flowers of sulphur are placed on the 
surface of bile they slowly sink to the bottom, but with water only the surface tension 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 3). 42 



326 PROFESSOR FRANK J. COLE 

is sufficiently high to keep the sulphur at the surface. The rate of fall of the sulphur 
particles was much more rapid with ox bile, but nevertheless with Myxine bile quite 
a large quantity of the sulphur sank to the bottom of the tube. 

Since the above was written, Professor Mekk obtained a fresh supply of Myxine, 
and I at once went to Cullercoats and collected the bile from about 100 specimens. 
I find the best way to do this is to remove the two lobes of the liver separately, and 
then detach the gall bladder from the gut. Severing the hepatic ducts and the bile 
duct does not cause bile to escape from the cyst. The latter can then be washed free 
of blood, and slit open inside a test-tube. All the bile is thus collected without any 
admixture of blood. It varies greatly in colour from a pale yellow to a deep green. 
The colour of the liver also varies considerably. 

The fresh bile is rather more gelatinous than that taken from animals kept for some 
time in the laboratory tanks, and consequently the precipitate of nucleo-protein obtained 
by acidifying the bile with acetic acid is much more pronounced. In the ether and 
hydrochloric acid test a much greater proportion of pigment was extracted by the use 
of dilute acid. 

Gmelin's ring test was more successful with the fresh bile. This is better carried 
out by pouring a thin film of bile on to a clean porcelain tile, and then adding the acid. 
The rings visible were a very faint yellow, red, violet passing into blue, and a very 
faint green. The red and blue zones were the most predominant, and this suggests 
that bilicyanin is in excess of the other pigments. 

4. Histology of the Gall Bladder. 

The gall bladder is lined by a single layer of epithelium. The cells are tall and 
narrow, and may measure as much as 35 m by 3 m- The nucleus is at the base and the 
chromatic material very concentrated, so that the nucleus stains a solid black with 
iron hsematoxylin. The free border of the cells, looking like a string of beads, is 
denser than the remainder, and readily strips off in imperfectly preserved material, or 
the free surface may dispatch into the lumen of the gall bladder innumerable sessile 
droplets. The cells are not in apparent contact, but are separated by a narrow non- 
staining zone. In badly preserved material the cells shrink considerably, and are then 
separated by intervals wide enough to be distinguishable under the low power of the 
microscope. The cytoplasm of the cells is coarsely vacuolated. 

Immediately investing the lining epithelium is a complex fibrous coat. This is 
essentially the vascular investment of the gall bladder, which, as shown by injections, 
is richly vascular with a well- developed capillary system. Arteries and veins and their 
connecting capillaries are readily distinguishable. Blood-vessels are very numerous 
just under the epithelium, although in nearly all cases a few fibres intervene between 
the vessels and the base of the epithelial cells. I have, however, here and there seen 
fine vessels penetrate between the lining cells themselves. 

The fibrous coat falls readily into three layers — a central transverse one and an 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 327 

outer and inner coat in which the course of the fibres is predominantly longitudinal. 
The central coat has a specific staining reaction, and in general appearance differs from 
the other two coats, its nuclei being far more numerous and up to 30 m long, and its 
fibres more accurately parallel. It does not appear in all cases to run continuously 
round the gall bladder, but is interrupted at intervals. It is the only one of the three 
coats which is distinctly muscular, the other two being composed of connective-tissue 
fibres. The inner coat is the most vascular. The connective-tissue fibres have shorter 
and more irregular nuclei, and have a coarse, granular, wavy appearance. In some 
methyl-blue-eosin preparations the connective-tissue nuclei stain blue, and the nuclei of 
the muscular layer red. In other cases the order of the layers appears to be reversed 
— the outer layers being muscular and the central one connective-tissue, and the 
conditions may even vary in different regions of the same gall bladder. 

A number of measurements of the thickness of the various coats of the gall bladder 
results in the following average : — epithelium, 26 /* ; inner fibrous coat, 29 m ; middle 
coat, 16 m; outer coat (very variable), 41 fx; total thickness of the gall bladder, 
112 Ai. These measurements apply to the right lateral free surface of the gall bladder. 
On the left the outer coat passes externally into a very loose connective-tissue layer 
transmitting the larger blood-vessels. 

Maas has a few notes on the histology of the gall bladder in his work on the 
"pancreas" of Myxine (p. 6). He finds in "old specimens and sections in a favour- 
able plane " two layers of unstriped muscle — one circular and one longitudinal. The 
layer he misses is evidently the outer one, which I also have found sometimes 
represented by a wide zone of very loose connective tissue. His description of the 
other two as both muscular is doubtless well within the variability of the animal. 

5. Histology of the Hepatic Ducts. 

The hepatic ducts consist of a central single layer of epithelium, without a basement 
membrane, almost directly under which is a rich vascular network. The epithelium 
is surrounded by a coat of coarse connective-tissue fibres having an average thickness 
of 60 /«, and which lodges the nerve bundles. This itself is enclosed by a very loose 
connective-tissue sheath. When the hepatic duct is cut across a triple lumen is 
disclosed. The largest is that of the duct itself ; the next is one of the two branches 
of the common portal vein ; and the smallest is a branch of the cceliac artery. In the 
neighbourhood of the artery and vein is found also an accompanying nerve. Further 
smaller vessels and nerves are present, as above mentioned. 

The hepatic duct is circular in transverse section and measures about 300 /j. in 
diameter in an average specimen. Its lumen is not regular, but is indented here and 
there. The lining epithelium consists of tall narrow cells about 72 m high and 4 m 
wide. The nuclei are at the base, and are in marked contrast with those of the gall 
bladder. They are oval and vesicular, with a well-marked nuclear membrane and 
reticular chromatin. The cytoplasm is granular and may be vacuolated, the cell 



328 PKOFESS01{ FRANK J. COLE 

outlines are not very clearly defined, and the free border is somewhat modified and 
tends to flake off" into the lumen of the duct. Occasionally an isolated nucleus in 
mitosis may be seen between the row of basal nuclei and the free border. 

The wall of the gall bladder is not perforated by the hepatic ducts at right angles 
to its surface, but very obliquely, with the result that a flap is formed which acts as a 
valve to the opening. At this point the transition from the epithelium of the hepatic 
duct to that of the gall bladder may be well studied. The cells of the duct become 
shallower and vacuolated, the non- staining areas between the cells appear, the 
modification of the free border becomes emphasised into the moniliform strip of the 
gall bladder epithelium, and the clear vesicular nuclei pass into the solid irregular nuclei 
of the bladder. On the side of the valve the transition from one kind of nucleus to 
the other is more sudden. 

6. Histology of the Bile Duct. 

The bile duct, like the hepatic ducts, consists of a single layer of irregular epithelium 
exhibiting numerous mitoses and without a basement membrane, and surrounded by 
connective-tissue coats. There are, however, some obvious differences. It is, of course, 
larger, and has a diameter of 525 m in a moderate-sized specimen outside the region of 
the "pancreas." The lumen accounts for at most 225 m of this. The wall is also 
somewhat thicker than that of the hepatic duct, but, owing to the pitting of the 
epithelial lining, this varies very greatly. In an average case the epithelium is 105 /« 
tall, and the connective-tissue coat 75 /«. The bile duct perforates the gall bladder 
more or less obliquely, but the aperture appears to be permanently and widely open, 
since I find no signs either of a sphincter or a valve. The transition of the epithelium 
of the bile duct into that of the gall bladder resembles that elsewhere described in the 
case of the hcDatic ducts. 

J. 

As the bile duct approaches the gut its wall becomes thickened owing to the 
presence of the "pancreas," the acini of which are situated within the thickness of the 
connective-tissue sheath of the bile duct. The latter penetrates the wall of the gut 
almost at right angles, and has a large opening at the apex of a prominent papilla, the 
connective-tissue coats being here of greater thickness. At the base of this papilla 
the lumen narrows down somewhat, but so far I have not found more than a sparse 
collection of unstriped muscle fibre in connection with the opening into the gut, and in 
fact there may be only equivocal traces of unstriped muscle in any part of the bile duct. 
In other examples, however, an unstriped layer immediately external to the epithelium 
is plainly establishable. 

The lumen of the bile duct is not regular, but is thrown into many pits or bags. 
This is due both to the folding of the epithelium, and also, and rather, to the variability 
in height of the cells themselves. Many of the pits are long and narrow, as if they 
were the terminations of glandular ducts ; but in nearly all cases they end blindly, in 
a few cases in an ampulliform enlargement, without reaching the periphery of the 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 329 

epithelium. Again, narrow duct-like or vesicular cavities appear quite frequently in 
the epithelium, and terminate blindly in both directions. Further, I have seen a duct 
bent like a U and opening at both ends into the lumen of the bile duct. Again, they 
appear quite freely in parts of the bile duct outside the region of the supposed pancreas. 

The appearance presented by many of the vesicular cavities within the epithelium 
of the bile duct is that of abortive evaginations. The cavity is surrounded by a layer 
of non-glandular epithelial cells resembling those of the bile duct epithelium, and it 
may or may not communicate with the lumen of the bile duct. Such structures are 
commonly present in the centre of the lobules of the "pancreas." They give to those 
parts of the bile duct where they occur the appearance of an epithelium of more than 
one layer, and in any case it is not strictly unilaminar, as, for example, are the hepatic 
ducts and the mucous membrane of the gut in this region. 

The free surface of the epithelium has a sharp cuticular edge, according to Maas, 
and flakes off into the cavity of the duct. The cells also extrude a substance which 
stains black with iron hsematoxylin. This material may in some cases be seen within 
the cytoplasm of the epithelium, whilst the lumen of the duct itself is devoid of it. 

The nuclei of the epithelium of the bile duct are oval, clear, and vesicular, and 
exhibit a reticular chromatin sometimes with, but often without, definite nucleoli. The 
cells themselves have well-defined cell walls. 

7. The " Pancreas." 

In 1896 (37 and 38) Maas described what he called a pancreas-like organ in Myxine 
and Bdellostoma/" He found, surrounding a portion of the bile duct, and also partly 
embedded in the serosa of the gut, and somewhat more strongly developed in older 
specimens, a peculiar glandular organ which he interprets provisionally as a pancreas. 
He states that it may be recognised macroscopically, after the loose surrounding tissue 
has been cleared away, as a yellowish white, lobulated envelope, which renders obscure 
the course of the bile duct, and which cannot be removed without damage to the duct 
and gut. It lies not symmetrically round the duct, but mostly on the left side, thus 
diminishing on the side of the gall bladder. Near the opening of the bile duct into 
the gut, however, the glandular mass forms a ring round the duct about equally 
developed on all sides. It consists of a number of lobules, each of which is stated to 
have a cavity in communication with the lumen of the bile duct. The secretion, then, 
whatever its nature, would be discharged into the bile duct. 

The following notes must be regarded, not as a complete description of the 
" pancreas " of Myxine, but as supplementing the descriptions of Maas. 

Counting the papilla within the lumen of the gut, the bile duct, from gall bladder 

* It seems quite possible that the structure described by Maas had already been discovered by Schneider in 
1879. He says (49, pp. 95-6): "Uin die Miindung [of the bile duct in Myxine] liegen eine grosse Zahl von 
Follikeln iihnlich wie bei Ammocotes." There is nothing but the "pancreas" in this neighbourhood to explain such 
a passage. Maas quotes Schneider's work, but appears to have overlooked his statements relating to Myxine. 



330 PEOFESSOR FRANK J. COLE 

to gut, had a total length in a 31 -cm. Hag of less than 4 mm. Of this the pancreas 
extended over rather less than the half nearer the gut. 

The pancreatic alveoli, the structure of which varies very considerably, are situated 
within the thickness of the connective-tissue coat of the bile duct, and in fact largely 
within the serous wall of the gut. They may be found also even at the apex of the 
papilla on which the bile duct discharges. Usually, however, there are more connective- 
tissue fibres externally than internally to the alveoli, although I have always found a 
well-marked zone of connective-tissue fibres between the pancreas and the epithelium 
of the bile duct. 

The pancreas consists of a number of lobules or alveoli clustered round the distal 
half of the bile duct. There is a vascular plexus round the bile duct, which also passes 
freely between the lobules of the pancreas, the vessels, however, always remaining 
outside the periphery of the lobules and never penetrating within them. The whole 
gland is surrounded by a strong sheath of connective tissue, which further extends 
freely among the alveoli, separating them from each other. There is no basement 
membrane to the alveoli, but each is separated from the connective-tissue framework 
by a very sparse layer of unstriped muscle. The lobules vary greatly in size, a fairly 
large example having a diameter of 255 a. Most, but not all, of them have a cavity, 
generally a relatively large cavity, which may have a diameter of even 195 m. One 
specially large alveolus had a diameter of 345 /j- and a cavity of 240 m. 

The development of the lobular cavities varies considerably.* In those cases where 
they are very large and frequent, I still do not find any connection between them and 
the lumen of the bile duct. Sometimes the cavities of adjacent lobules approximate 
and are only separated by a narrow partition. Nevertheless I have observed no fusion. 
Again, spaces may appear sporadically in places outside the region of the pancreas, as, 
for example, at the gall-bladder end of the bile duct, within the lining epithelium of the 
duct itself. Instead of one large cavity a lobule may have five or more smaller ones, 
which appear and disappear in an apparently capricious manner without discharging 
anywhere. In nearly all cases the cavities are either partly or wholly surrounded by a 
single layer of epithelium showing numerous mitoses, as in the epithelium of the bile 
duct, and having in general a close resemblance to the latter epithelium. It is, in fact, 
difficult to escape the conclusion that the pancreatic lobules are outgrowths from the 
bile duct. 

Very rarely the connective tissue separating an alveolus from the lining of the bile 
duct is interrupted, and the glandular mass fuses with the biliary epithelium, but only 
in a few isolated cases have I observed any connection, or any appearance which might 
be interpreted as. a connection, between the cavity of the gland and the lumen of the bile 
duct. I have even seen the secretory section of a gland fuse with the epithelium of 

* Since tlie following paragraph was written 1 huve examined another series of sections, in which all the pancreatic 
follicles were homogeneous and solid, and in which there conld be no ([uestion whatever of any dncts connecting them 
with the bile duct. Fusions between certain of the follicles and the epithelium of the bile duct were observed, but 
these fusions were wilhout the .slightest trace of a cavity. 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 331 

the bile duct — without any connection of the two cavities — whilst the duct portion of 
the same gland was at the opposite end and turned away from the bile duct. In fact, I 
find it to be the rule rather than the exception for the duct portions of the lobules to 
be directed away from the cavity into which they are supposed to open. Indeed, in 
practically all my preparations the pancreas is a ductless gland, although in saying this 
I do not wish to claim that such a condition is any other than a secondary and 
acquired one. The lobules with the largest cavities are often those which are absolutely 
and indubitably shut otf from any connection with the bile duct. Owing to the strong 
selective staining of the connective-tissue framework, there can be no question here of 
any error of observation. Further, I have never found that injections thrown into the 
lumen of the bile duct pass into the cavities of the pancreatic follicles. If the latter 
were in connection with the lumen of the bile duct, such connections could easily be 
observed where the connective-tissue sheath is penetrated. In the very few cases where 
the connection was found, it was easily traced even with the low power of the microscope. 

Several pancreatic lobules may fuse and have a perfectly defined epithelial duct 
with a diameter of 144 m, and formed by a single layer of epithelium 36 m high, enclos- 
ing a cavity 72 /x across. Such a duct may lose and acquire its lumen in a very curious 
manner, and finally terminate in a cul-de-sac quite apart from the lumen of the 
bile duct. 

Usually the lobules are not isolated, but dichotomise so as to form connected groups 
after the manner of an acinous gland. Some, however, may be quite independent. 
There is, in fact, an absence of structural stability one does not usually find in an organ 
of functional importance. The pancreas occurs only on one side of the bile duct, until 
shortly before the latter opens into the gut, when it completely surrounds the bile duct 
like a ring. The alveoli contain sometimes a colourless liquid, and sometimes a granular 
substance including variously-sized particles — some of them quite large, and others 
having a concentric appearance — staining an intense black with iron hsematoxylin. 
Here and there one finds a single disintegrating cell in the alveolus. 

The glandular as distinct from the epithelial portion of the lobule has a somewhat 
complex structure. The nuclei are oval or round and have precisely the same appear- 
ance as those of the duct portion of the lobule and of the bile duct itself, except that 
the latter are larger and the chromatic network is somewhat looser and therefore stains 
more lightly. In some cases a well-defined nucleolus can be seen lying in a clear area. 
Very few mitoses were, however, observed. The cell outlines are sometimes difficult, or 
even impossible, to define. In other cases they appear quite clearly, and the gland cells 
are then seen to be irregular in shape and to vary in size. The cytoplasm is a loose, 
faintly staining reticulum, and not granular. It may contain large non-staining spaces. 
Inter- or intra-cellular cavities as described by Maas I have not observed ; but these, 
like bile capillaries, may depend on the state of activity of the gland. A tangential 
section of a lobule shows what might be interpreted as inter-cellular cavities, but these 
on examination turn out to be the inter- lobular blood capillaries. I am speaking now of 



332 PROFESSOR FRANK J. COL?] 

material specially fixed by intra-intani injection, in which I do not find any specially 
modified cells associated with the possible existence of inter- or intra-cellular spaces. 
Further, the gland cells of the pancreas have certainly not the appearance of an enzyme- 
secreting gland, but if anything that of a mucin gland. 

The fundamental point of difference between Ma as' results and my own is that he 
describes the secretion of the gland, whatever it may be, as discharged freely into the 
bile duct, whilst in my material the gland is to all intents and purposes a ductless one. 
I have no doubt we have both correctly described the material at our disposal as we 
found it, and that the difference in our results is only one more indication of the exist- 
ence of distinct races or varieties of Myxine glutinosa. \i is clear that the pancreatic 
lobules represent outgrowths of the bile duct, and are hence, at least originally, in free 
communication with it. It seems therefore that the absence of the efferent channels in 
my material indicates a retrogressive change, and that the gland is either losing the 
importance it formerly possessed, or undergoing some functional modification. 

Owing to the small size of the gland, and the great difficulty of satisfactorily isolating 
its secretion, I do not think it possible to test the latter by biochemical methods. It would 
mean dissecting it out in at least 100 specimens ; and even if this could be done in the 
necessary time, one could never be certain that the extract was free from the secretion 
of the liver. 

H. The Thyroid Gland (Figs. 3 and 12). 

In a preliminary paper (13) I briefly described the structure of the thyroid of 
Myxine, under the impression that it had up to that time escaped notice. Schaffer, 
however, pointed out that the Myxinoid thyroid had been described by W. Muller in 
an obscure paper in 1871, and at the same time he added an account of this structure 
in Myxine. I shall, therefore, now confine myself to such details as are omitted in 
Schaffer's description. 

As shown in fig. 1 2, which represents a transverse section through the region of the 
gills, there is a column of loose fatty tissue in the middle line, wedged in between the 
gills, and extending from the gut above to the cardiac aorta below. In this column of 
fatty tissue are embedded the small vesicles which represent an unpaired, diffuse, 
scattered thyroid, and which can be seen with the naked eye {thy.). 

The thyroid at first sight does not appear to be an important organ in the adult. 
In fig. 12, which in this particular series showed the maximum number of thyroid 
vesicles appearing in any one section, only eight are seen, and the gland certainly does 
not strike one from the point of view of size. Fortunately, however, I plotted it out 
from my large series of sections, and the result is given in fig. 3. Nothing could be 
more striking. The gland extends over a considerable area, both vertically and 
longitudinally, and it includes in this individual 251 vesicles of all sizes and shapes. 
If these were massed together into one compact gland, we should have, compared with 
the size of the animal, a large and obviously important structure. 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 333 

All the vesicles are closed and have no connection with each other, and, as I pointed 
out in my preliminary paper, the thyroid of Myxine in this respect resembles that 
of Teleosts. Schaffer has found, exceptionally, some of the alveoli dorsal to the 
oesophagus, which is contrary to my experience. In my large series of sections the 
vesicles commence somewhat in advance of the first pair of gills at section 2096 
(cp. the chart, fig. 3). These anterior vesicles are flattened dorso-ventrally, and are 
therefore wide from side to side, and although the wall consists of the usual thyroid 
epithelium, only a very slight cavity and contents were observable in the sections. In 
fact, the anterior vesicles are difficult to find, and can hardly be important functionally. 
They are closely attached to the thick tough membrane which here binds together the 
copulo-copularis and perpendicular muscles and the two chondroidal bars, and they are 
situated immediately over the superior chondroidal bar. The most posterior vesicles do 
not extend much behind the origin of the last afferent branchial artery, and not so far 
back as the posterior border of the last gill. At this point the gut becomes deeper 
dorso-ventrally, and descends to enter the anterior coelom. 

The form and size of the vesicles vary enormously, although most of them are 
small and circular or oval. In the specimen figured the large and irregular vesicles 
were confined to the ventral region of the thyroid area, but this may not apply to all 
cases. Schaffer does not appear to have realised the true form of the alveoli, having 
evidently only studied detached sections without reconstructing entire alveoli. 

Admirable preparations illustrating the histology of the thyroid may be obtained by 
intra-vitam injection of the fixative from the heart. I have used mostly Mann's picro- 
corrosive-formalin. This travels through the vessels with great rapidity — much faster 
than any coloured injection medium, and the whole animal is fixed throughout in a 
fraction of a minute, as can be demonstrated by opening up the body in various places 
immediately afterwards. I find the best stain for Myxine tissues generally is iron alum 
hsematoxylin, followed by eosin. 

The wall of the thyroid vesicles consists of a single layer of epithelium, in average 
cases about 20 m high. This may be quite flat and simple in structure, or the cells 
may be fairly tall and with well-marked intra-cellular contents. Generally they ex- 
hibit a well-marked resemblance to the thyroid vesicles of other vertebrates. There 
is, as in man, no basement membrane sensu stricto. It is true the vesicles are enclosed 
in a sheath, but this is formed by, and belongs to, the surrounding connective tissues. 
That it is no part of the vesicle is well seen where two vesicles are in close contact. 
The supposed basement membrane passes across from one vesicle to the other, leaving 
the contiguous surfaces bare. Or in those cases where a vesicle is closely opposed to 
the ventral wall of the gut, it will be noticed that the abutting surface of the vesicle is 
without a membrane. 

The histology of the thyroid cell varies very greatly according to its condition of 
activity. The nuclei are well marked and vesicular. There is a distinct nuclear mem- 
brane, and a reticulum with a somewhat dotted chromatic material and nucleoli. Apart 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 3). 43 



334 PKOFESSOR FRANK J. COLE 

from the nucleus, the contents of the cell vary considerably, not only in appearance but in 
staining reactions. Sometimes we can distinguish intra-cellular bodies of two kinds, which 
may, however, both occur in the same cell. First, there are large waxy bodies, which 
vary in consistency, and lie in a clear area or vacuole. In their early (?) stages these 
stain faintly with eosin and iron hajmatoxylin. Then there are the smaller spherical 
bodies, which may be very numerous and variable in size, and appear to resemble the 
zymogen granules of other secreting cells. They stain intensely with iron hsematoxylin 
and eosin. The former have been described by Schaffer, but they are not invariably 
present, whilst the latter he does not mention. In some cases where these two intra- 
cellular products are extremely developed, the nuclei are indistinct, and only stain 
faintly. There are many indications that the larger intra-cellular bodies are formed by 
the fusion of the smaller ones. 

When there is a single large intra-cellular product present it lies, as a rule, either on 
the peripheral or the central side of the nucleus. Schaffer states that it occurs only 
rarely at the peripheral end of the cell, but this statement appears to be based on the 
examination of an insufficient number of individuals, since it may be found more often 
peripherally than centrally, and in some cases I have found it almost exclusively on 
the outer side. In a few cases there are two products in one cell, which may be either 
together central or external to the nucleus, or separated, with the nucleus crushed 
between them. These products in isolated cases may be extruded bodily from the 
cell into the cavity of the vesicle. 

Schaffer figures one condition which I have not seen, although I do not on that 
account question the accuracy of his observation. Here the nucleus is pushed to the 
periphery of the cell, and is greatly compressed from side to side so as to assume a rod- 
like form, whilst the central portion of the cell is occupied by a large, oval, feebly 
staining secretion product, surrounded by a sheath or " theca." 

The contents of the vesicles are again very variable. Usually they appear quite 
empty in the sections. Occasionally one finds a coagulated substance of a granular 
appearance, and only very rarely, as Schaffer also finds, is there present the colloidal 
substance so characteristic of the thyroid vesicles of higher vertebrates. 

The thyroid of Myxine was first described by W. Muller in 1871 (43) in the 
following words : " Dagegen ist mir der Nachweis des Organs in der Classe der Cyklosto- 
men bei Myxine glutinosa gelungen, welcher die Schilddriise bisher allgemein abge- 
sprochen worden ist. Sie liegt hier in der fettreichen Bindegewebslamelle, welche sich 
von der Ventralfiache des (Esophagus zur oberen Flache des Kiemenarterienstammes in 
dessen ganzer Ausdehnung erstreckt und besteht aus einer ziemlich betrachtlichen Zahl 
theils zerstreut liegender isolirter, theils zu kleinen Gruppen von 2-5 vereinigter rings 
geschlossener Follikel, Letztere sind theils von kugeliger, theils von ellipsoidischer 
Gestalt, der Durchmesser der ersteren schwankt zwischen 0,1 und 0,25 der Liingen- 
durchmesser dur letzteren erhebt sich bis zu 0,4. Sie bestehen aus einer dlinnen Mem- 
brana propria und dieser aufsitzendem einschichtigem Epithel. Die Zellen des letzteren 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 335 

sind theils cubisch, 0,008 im Durchmesser, theils cylindrisch, 0,012 hoch, 0,008 breit, 
sammtlich mit Kernen von durchschnittlich 0,006 und 1-2 Kernkcirperchen und sehr 
zartem, feinkornigem Protoplasmakorper versehen. Das Epithel umschliesst eine scharf 
begrenzte rait klarer, farbloser Fliissigkeit gefiillte Hohle." This description, though 
brief, admits of no doubt as to Muller having seen the thyroid of Myxine. The same 
writer a year later* describes the thyroid of the adult lamprey as extending below the 
longitudinal " tongue " muscle from the second to the fourth pair of gill sacs. 

The development of the Myxinoid thyroid has been studied by Stockard in 
Bdellostoma (5S). He finds that it arises as a median, unpaired " downpushing from 
the ventral floor of the pharynx throughout the entire gill area." This forms a long 
trough with a restricted lumen, which subsequently becomes separated from the gut, 
and breaks up to form the isolated vesicles of the adult. " The chief point of interest 
shown by the trough-like thyroid anlage is the very extensive gut area from which it 
is derived ; in no other vertebrate does the thyroid evagination from the pharynx run 
through so relatively long an area " (p. 95). " It will now be clearly seen that although 
ihe adult condition of the thyroid in Myxinoids and Teleosts are readily comparable, 
the developmental processes through which the ends are reached seem widely different " 
(pp. 97-8). Judging from the recent work of Ferguson and Gudernatsch, the thyroid 
is a remarkably uniform structure throughout the Chordate series. 

J. Cloaca (Fig. 5). 

The cloacal aperture is an elongated longitudinal slit compressed from side to side, 
9 mm. long in a 36-cm. Hag, and 4|- cm. from the extremity of the tail. The anatomy 
of the cloacal region was first fully described by R. H. Burne (12). If an incision be 
made from the anterior extremity of the cloacal aperture and continued a short distance 
along the rectum, and the walls pinned out, it will be noted that the anus (an.) opens 
into the cloaca (cl.) anteriorly and ventrally by a larger puckered aperture. The roof 
of the rectum [an.') is continued backwards across the cloaca at the sides, but not at 
the middle line, and thus the cloaca is divided imperfectly into a dorsal and a ventral 
chamber. Both chambers bear folds. In the one case they are continued backwards 
from the margin of the genital pore, and in the other they are the direct continuations 
of the rectal folds. The ventral chamber receives the anus, as already mentioned. 
Into the dorsal chamber there opens in front, and of course dorsal to the anus, the 
single, conspicuous, round abdominal pore, or porus genitalis (p.g.), surrounded by a 
thick fibrous band {p.g.'). This pore was first described by A. M. C. Ddmeril in 1807, 
in Myxine, and afterwards by J. Mijller. From the porus genitalis there passes back- 
wards right to the posterior end of the cloaca a narrow but obvious median dorsal ridge. 
Somewhat in front of the middle of this ridge there is a thickening, the urinary papilla 
{ur.pp.), and an examination of this thickening with a lens reveals two small 

* Jena. Zeits., Bd. vii., 1872. 



336 PROFESSOR FRANK J. COLE 

asymmetrical openings — the apertures of the segmental ducts (s.d.) The left seems to 
be the larger, and is certainly anterior to the right. There thus appears to be in 
Myxine, in this dorsal cloacal chamber, a representative of the distinct and undoubted 
urogenital sinus of the lamprey, but Burne argues with much reason " that the uro- 
genital sinus of the lampreys is absent in the Myxinoids, and that in the latter the anus, 
' porus genitalis,' and ureters open into an integumentary cloacal chamber, similar to the 
cloacal chamber common to anus and uro-genital sinus in the lamprey " (12, pp. 494-5). 

According to Burne, the porus genitalis is surrounded by a large diffuse gland, the 
histology of which is said to make it highly probable, in spite of its deep position, that 
it represents modified lateral slime glands. As v/ill be seen on reference to fig. 5, 
the slime glands {s.s.) are apparently absent from the two myotomes (89, 90) of the 
cloacal region, and when it is remembered that the supposed genital pore gland 
completes the only break in this linear series, and that the true slime glands immediately 
posterior to it are also largely covered by the caudal muscles (cp. Part II., fig. 4), this 
conclusion receives some support. The gland is stated to have several openings in the 
neighbourhood of the margin of the pore. Its function may be, as Burne suggests, 
either to lubricate the pore during oviposition, or to occlude it by a waxy secretion, 
and thus prevent access of foreign matter to the abdominal cavity. These suggestions, 
however, disregard differences which should exist in relation to sex, and also the fact 
that the genital pore is closed by the action of the sphincter cloaca\ I shall discuss 
the alleged anal slime gland in a subsequent passage. 

The hind-gut has only a short course. It is easily distinguished from the mid-gut 
by its sharp, acute, and somewhat irregular folds, which have no lymphoid packing, 
and are similar to those of the fore-gut in having an epithelial investment of several 
layers of cells. The lining of the cloaca even more resembles the epithelium of the 
outer skin. 

I have already drawn attention (cp. Part I., p. 786) to the thin sheet of cartilage 
described by Ayers and Jackson in the wall of the cloaca of Bdellostoma. I have 
described in the section on the gut the only representative of cartilaginous tissue which 
occurs in the cloacal region of Myxine. 

An original figure of the cloacal region and tail of Myxine glutinosa is given by 
Goodrich in his masterly treatise on Fishes in Lankester's Zoology (p. 33). 

Since the above was written I have made and examined serial sections of the cloaca 
of a 19-cm. and a 25-cm. Hag. In both cases the segmental duct expanded posteriorly 
into a kind of sinus. From this expansion, which ends blindly behind, there is given 
off internally another and terminal duct,* which at once bends mesially, dips urider 
the sphincter cloacae, and, passing downwards and backwards, opens at the urinary 
papilla as above described. Now the place where this terminal duct is given off 
coincides precisely with the boundary between the mid- and hind-guts, and it is 
therefore highly interesting to note : ( 1 ) that the lining of the segmental duct consists 

* A somewliat similar condition is figured by Burne in Bdellostoma. 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 337 

of a single layer of cells, but that of the terminal duct of many layers of cells ; (2) 
that the two epithelia bevel into each other exactly as they do at the mid- and 
hind-gut boundary (cp. description of gut) ; and (3) that the glassy and granular 
mucous cells characteristic of the epidermis may occur at any part of the terminal 
duct. In the 19 -cm. Hag the two terminal ducts open separately and symmetrically, 
but in the 25-cm. specimen the openings were asymmetrical, as already described in 
the mature animal. It seems natural to conclude that the segmental duct terminates 
at the anterior border of the constrictor cloacae in the sinus-like swelling, and that the 
terminal duct is simply an integumentary (epidermal) pocket. It must, however, not 
be forgotten that on similar grounds we should have to regard the whole of the 
pharyngeal gut as epidermal, and this we know to be contrary to fact. 

In one respect the description given above is directly contrary to the following 
statement by Burne. He says : " Now in the adult myxinoid the ureters do not 
imperceptibly pass into the cloacal chamber, as they do into the urogenital sinus of 
the lamprey, but open upon a raised papilla ; upon the margin of the ureteric opening, 
the epithelium changes its character — inside, it is sirmlar to that lining the rest 
of the ureter; outside, it is epidermic'' [italics m.ine]. 

In the transverse sections of the 19-cm. Hag the rectum fuses ventrally with the 
abdominal wall at the anterior ventral margin of the sphincter cloacae, and further 
back the whole fuses with the skin at the boundary between mid- and hind-guts. 
The gut is now connected with the roof of the abdominal cavity by the deep median 
dorsal mesentery. Then the striated sphincter cloacas rises round and over the roof 
of the gut and splits the body cavity into two portions, placed one over the other, and 
each divided by a median vertical partition — the posterior continuation of the 
mesentery. The dorsal pair of cavities soon die away, the left one first. The ventral 
pair one naturally expected to open posteriorly at the genital pore. The median 
mesenteric partition, first of all, becomes imperfect ventrally, but the dorsal portion 
remains suspended from the roof of the cavity almost to its posterior extremity. The 
cavity becomes narrowed and surrounded by a dense cellular connective tissue, which 
finally fuses with the upper section of the cloaca, but the lumen ends blindly ivithout 
opening anywhere. There was, therefore, no porus genitalis in this specimen, nor the 
slightest trace of the anal slime sac described by Burne. In the series of longitudinal 
sections of the 25-cm. Hag the constrictor cloacae does not partition the body cavity 
as above described, but passes entirely above it. The posterior blind extremity of the 
body cavity bends downwards and backwards, and approaches a blind epidermal cloacal 
pocket. The two are separated by a plug of cellular connective tissue. A close 
examination of this plug reveals phagocytic operations in its central core, and there is 
no doubt that in this specimen the genital pore was in the act of breaking through. 
This, of course, may happen at diff"erent sizes according to sex and conditions of 
maturity. In this example also there were no signs of the anal slime sac. Hence the 
genital or abdominal pore of Myxine is to be found only in the mature animal. It 



338 PROFESSOR FRANK J. COLE 

should be noted that the constrictor cloacae, by enclosing the anus, genital pore, and 
segmental ducts, will occlude all three at the same time. 

To add a still further note to the above, 1 have now, by the courtesy of Mr E. H. 
BuRNE, examined his sections of the cloaca of Myxine. I find in the ureter that 
the change from the single- to the many-layered epithelium occurs precisely as in my 
own sections. The statement by Mr Burne, quoted above, that the epithelium changes 
at the margin of the ureteric opening, is therefore incorrect. The only difference of 
importance is that in Mr Burne's sections I find no glassy or granular mucous cells in 
the terminal duct. These are, however, not numerous in my sections, and they may, 
of course, disappear in the adult. With regard to the anal slime sac, Mr Burne has 
been somewhat, and naturally, misled by not having examined sections of younger 
material. The structure in question is undoubtedly what I have called above the 
dorsal chamber of the cloaca, into which the genital pore opens, and the " semicircle 
of openings " of the anal slime sac are simply the spaces between the folds into which 
the lining of the cloaca generally is thrown. The dorsal chamber is naturally present 
in my sections ; but, although it is crowded with the glassy and granular mucous cells, 
it never occurred to me to identify it with Mr Burne's anal slime sac, since these cells 
occur just as numerously in another region of the cloaca. I can, however, confirm 
Mr Burne in one important respect, i.e. in the presence of the peculiar thread cells 
in the dorsal chamber of the cloaca which are found elsewhere only in the lateral slime 
sacs of the skin. Also the epithelium is crowded with the glassy raucous cells, but 
there are no granular cells. In my preparations there are numerous granular cells, 
but no thread cells. The slime sac character of the dorsal chamber of the cloaca 
therefore develops late, and there is little doubt that the thread cells are produced by 
modification of the granular cells. There is no occasion to conclude that the dorsal 
chamber is a modified slime gland or glands. The whole skin produces glassy mucous 
cells and thread cells of a type, as shown by G. Retzius, and the fact that the 
slime sac characters have not been fully assumed in a 25-cm. Hag would indicate that 
the conversion of the dorsal chamber of the cloaca into a kind of slime sac is a new and 
independent feature. 



K. LITERATURE. 



[Li tliis list only those works are included which contain original observations on the viscera 

of Myxinoids.] 

(1) ABiLDGAAun, 1792. Schrift. d. Gea. nat. Freunde z. Berlin, Bd. x. pp. 193-200, Tab. IV. 

Naso-palatine opening, respiratory apparatus, gall bladder and liver, and mesentery of Myxine. 

(2) Allis, 1903. Anat. Anz., Bd. xxiii. p. 333. 

Gill clefts of Bdellostoma. 

(3) Aykrh, 1894. Biological Lectures delivered at Wood's Holt, 1893, vol. ii. pp. 137 and 152. 

Number and variation of Gills in Bdellostoma. 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 339 

(4) Ayers, 1907. Lancet-Clinic, Dec. 28. 

Naso-hypophysial canal of Bdellostoma. 

(5) Aybrs and Jackson 1900-1. /tiz«-. MorpA., vol. xvii. p. 212 ff. Reprinted: Bull. Univ. Cincinnati, 

Ser. ii. vol. i. 

Gills and their variation in Bdellostoma. 

(6) Batbson, 1894. Materials for the Study of Variation, London, pp. 172-174. 

Variation in gills of Myxine and Bdellostoma. 

(7) Beard, 1888. Anat. Am., Bd. iii. p. 15. liejjrinted : Nature, vol. xxxvii. p. 224. 

Nasal duct of Myxine. 

(8) Beard, 1889. Zool. Jahrb., Abth. Anat., Bd. iii. p. 746. Preliminary notices : Nature, yoI. xxxvii. 

p. 499, 1888; and Anat. Anz., Bd. iii. p. 169, 1888. 
Myxinoid mouth neither suctorial nor glandular. 

(9) Blooh, 1793. Naturgeschichte der Ausldndischen Fische, Th. vii. p. 67, Berlin. 

Inaccurate observations on the breathing organs of Myxine. 

(10) Bloch, 1801. Sysi. Ichthyol., Ed. J. G. Schneider, Berlin. 

Respiratory organs of Myxinoids, p. 104. 

(11) Braus, 1896. Jena. Denkschr., Bd. v. (Semon, Zool. Forschungsreisen, ii.). 

Liver of Myxine. 

(12) Burne, 1898. Jour. Linn. Soc. Zool., vol. xxvi. p. 487. 

Cloacal region Myxine and Bdellostoma. 

(13) Cole, 1905. Anat. Anz., Bd. xxvii. p. 323. 

Notes on Myxine. 

(14) Cole, 1905. Trans. Roy. Soc. Edin., vol. xli. p, 749. 

Skeleton of Myxine. 

(15) OoiiB, 1907. Trans. Roy. Soc. Edin., vol. xlv. p. 683. 

Muscles of Myxine. 

(16) Cole, 1909. Trans. Roy Soc. Edin., vol. xlvi. p. 669. 

Skeleton of Myxine. 

(17) Cole, 1912. Trans. Roy. Soc. Edin., vol. xlviii. p. 215. 

Vascular system of Myxine. 

(18) Cunningham, 1886. Q.J. M.S., vol. xxvii. 

Nasal duct of Myxine. 

(19) Dean, 1895. Fishes, Living and Fossil. New York. 

General anatomy of Myxinoids. 

(20) Dean, 1899. Kupffer's Festschrift, p. 221. Cp. also : Science, N.S., v., 1897, p. 435, and ix., 1899, p. 311 ; 

Jour. R. Micros. Soc, 1898, p. 55, and 1900, p. 444; and Q.J. M.S., N.S., vol. xl. p. 269, 1897. 
Development of gills and gut of Bdellostoma. 

(21) DuMERiL, 1807. Mem. d'anat. comp. Anat. d. Lamproies. Paris. 

Gills of Myxine. Also first description of abdominal pore. 

(22) EwART, 1876. Jour. Anat. Phys., vol. x. p. 488. 

Cloacal region of Myxine. 

(23) FuRBBiNGER, 1897. Gegenbaur's Festschrift, Bd. iii. p. 620. 

Gills of Myxinoids. 

(24) Goodrich, 1909. Lankester's Treatise on Zoology, part ix. 

General anatomy of Myxinoids, with some original figures. 

(25) Gunner, 1763. Trotidhiemske Selskabs Skrifter, Dl. ii. p. 250. German translation : Drontheim 

Ges. Schrift., Th. ii. p. 230, 1765. 

Naso-pharyngeal duct (first description), gills, liver, and gall bladder of Myxine. 

(26) Haack, 1903. Zf.w.Z., Bd. Ixxvii. p. 112. Abstract : Jour. R. Micros. Soc, 1904, p. 170. 

Mouth and oesophagus of Myxine. 

(27) Holm, 1897. Zool. Jahrb., Abth. Anat., Bd. x. p. 277. Also pubhshed as Inaug. diss., Freiburg, 

i. B., 1897. 

Liver of Myxine. 



340 PROFESSOR FRANK J. COLE 

(28) Home, 1815. Phil. Trans., vol. cv. p. 256. Abstract: Abstracts Phil. Trans., vol. ii. p. 23, 1833. 

Reprinted in Home's Lectures on Comparative Anatomy, vol. iii. p. 165, and vol. iv., Pis. 46 
and 47, 1823. German translations: Meckel's Deut. Archiv, Bd. ii. p. 594, 1816, and Isis 
{Oken's Ency. Zeit), Bd. i. p. 25, 1817. 

Gills and respiration in Myxine and Bdellostoma. 

(29) Howes, 1893. P.Z.S., p. 730. 

Variations in gills of Myxine. 

(30) Jackson, 1901. Jour. Cincinnati Soc. Nat. Hist., vol. xx. p. 13. Reprinted : Bull. Univ. Cincinnati, 

Ser. ii., vol. i. Abstracts: Amer. Nat., vol. xxxvi. p. 338; and Jom?-. R. Micros. Soc, 1902, p. 174. 
Gills of Bdellostoma. 

(31) JoKDAN and Snyder, 1901. Proc. U. S. Nat. Mus., vol. xxiii, p. 730. 

Ductus oesopliago-cutaneus of Myxinoids. 

(32) Kalm, 1753. En Resa til Norm America, 3 vols., Stockholm, 1753-1761. T. I., p. 100. Swedish. 

Several English, French, and German translations published. 
First description of Myxine (German ed., vol. i. p. 1 18). 

(33) Klinckowstrom, 1890. Fdrhand. Biol. Foren. Stockholm, Bd. ii. p. 62. 

Gut and liver veins of Myxine. 

(34) KuPFFER, 1899. Sitz. Ges. Morph. Phys. Munchen, Bd. xv. p. 21. Issued separately, 1900. 

Development anterior region of gut of Bdellostoma. 

(35) Lebeboullet, 1838. Anat. Comj). appareil respiratoire d. I. animaux vertebres. Strassbourg. 

Variation in gills of Myxine (7 pairs). 

(36) Leydjg, 1857. Lehrbuch. d. Histologie. Hamm. 

Notes on histology of gut of Myxine. 

(37) Maas, 1896. Sitz. Ges. Morph. Phys. Munich, Bd. xii. p. 46. Review: Zool. Central, Bd. iii. p. 741. 

" Pancreas," bile duct, hepatic ducts, and gall bladder of Myxine. 

(38) Maas, 1896. Anat. Anz., Bd. xii. p. 570. Review: Zool. Central, Bd. iv. p. 361, 1897. 

"Pancreas" and bile duct of Bdellostoma. 

(39) Maas, 1899. Kupffer's Festschrift, p. 197. Review: Zool Central., Bd. vii. p. 525. 

General description of gut of Myxine. 

(40) MiJLLER, J., 1835-6. Abh. Ak. Berlin, 1834, p. 65, 1836. Published separately, Berlin, 1835. Cp. 

Wiegm. Arch., 1836, ii. p. 245. 

Respiratory organs of Myxine and Bdellostoma. Velum. 

(41) Mt)LLER, J., 1839, 184L 1842. Abh. Ak. Berlin, 1839, pp. 181-183, 1841. Published separately, 

Berlin, 1842. Abstract: Ber. Ak. Berlin, 1839. Cp. Miiller's Archiv, 1840. 
Structure of the Myxinoid gill. 

(42) MuLLER, J,, 1845. Abh. Ak. Berlin, 1843, p. 109, 1845. Published separately, Berlin, 1845. 

General description of the viscera of Mijxine and Bdellostom,a. 

(43) MuLLBR, W,, 1871. Jena. Zeits., Bd. vi. p. 433. 

Thyroid of Myxine. 

(44) Price, 1897. Sitz. math.-phys. A. k. buyer. Akad. Wiss. Munich, 1896, p. 69, 1897. Abstract: 

Verhand. anat. Ges. Berlin, 1896, p. 81. 

Development of mouth and gills of Bdellostoma. 

(45) Retzius, a., 1824. JC. vet. Acad. Hand. Stockholm, p. 408. German translation : Oken's Isis, 1825, 

Bd. ii. p. 1013. French translation : Ann. Sci. Nat., Bd. xiv. p. 148, 1828. 
Velum, gut, liver, and biliary apparatus of Myxine. 

(46) Retzius, G., 1892. Biol Unters., Bd. iv. p. 68. 

Bile capillaries and structure of liver in Myxine. 

(47) liETZius, G., 1895 Fries, Ekstrom, and Sundervall's History of Scandinavian Fishes, p. 1195. 

General anatomy of Myxine. 

(48) Schaffbr, 1906. Anat. Anz., Bd. xxviii. p. 65. 

Thyroid of Myxine. 

(49) Schneider, 1879. Beitr. z. vevjleich. Anat. u. Entwick. d. Wirbelthiere, Berlin. 

Gut of Myxine. 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 



341 



(50) ScHREiNBR, K. E., 1898. Berijen's Mus. Aarb., No. L Review: Zool. Central, Bd. vii. p. 66, 1898. 

Histology of gut of Myxine. 

(51) Stannius, ] 854. Zootomie d. Fische, 2nd ed. Berlin. 

General anatomy of Myxinoids. 

(52) Stockard, 1906. Amer. Jour. Anat., vol. v. p. 481. 

Development of mouth and gills of Bdellostoma. 

(53) Stockard, 1906. Anat. Anz., Bd. xxix. p. 91. 

Development of Thyroid of Bdellostoma. 

(54) Stockard, 1908. Anat. Rec, vol. ii. p. 336. 

Gill position of Myxinoids. 

(55) WiEDERSHBiM, 1883. Lelir. d. vergleich. Anat., pp. 590 and 603. 

Gut and biliary apparatus of Myxine. 

(56) WoRTHiNGTON, 1905. Ainer. Nat., vol. xxxix. p. 625. 

Gills of Bdellostoma. 



L. EXPLANATION OF PLATES. 



Rbpbrbnce Letters. 



an. 



a.p.v. 
a.s. 



b.d. 



anus 

two 



a.d.p. Anterior arch of the dental plate. 
a.h.d. Anterior hepatic duct. 
a.h.v. Anterior hepatic vein. 
a.l.l. Anterior lobe of the liver. 
an. Anus. 

Posterior extremity of roof of 
partially dividing cloaca into 
chambers. 
Anterior portal vein. 
"Aquseductus Sylvii," or forward 
prolongation of the mesencephalic 
ventricle. 
Bile duct. 
b.g. Biliary gland or "pancreas." 
h.o. Bulbus olfactorius. 

b.p. Pad of soft pseudo-cartilage at the 
anterior end of the anterior segment 
of the basal plate. 
b.p.^~^ Anterior, middle, and posterior segments 

of the basal plate. 
br.a.^ "First branchial arch." 
br.a."^ " Second branchial arch." 

7 // T^. •, f external branchial apertures. 
or.ap. Right J 

br.cl. Branchial cloaca. 
c.ao. Ventral or cardiac aorta. 
c.b.c' Paired anterior limb of the M. con- 
strictor branchiarum et cardise. 
Second loop of the same muscle. 
Third loop of the same muscle at the 

bend (cp. Part II., fig. 2). 
Cornual cartilage. 



c.b.c. 

c.b.c." 



e.c. 



c.e. 
e.e.' 
c.e." 



c.g.p. 



c.g.p. 

c.g.p." 

c.g.s. 

c.g.s.' 

c.g.if." 

c.h. 

cl. 

coe.a. 
cp.' 
c.p." 

cp.'" 
ep.c. 
e.pl. 

cp.t.c. 

cp.t.c' 
cp.t.c." 

c.p.v. 

c.q.p. 

c.q.s. 
d.m. 



of the M. 
moidalis. 



copulo - eth- 



TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 3). 



External head ^ 
Internal head 
Fused external 

and internal 

heads 
Tendon of the M. copulo-glossus 

profundus, the protractor muscle of 

the dental plate. 
Lateral head ) of the M. copulo-glossus 
Median head J profundus. 
M. copulo-glossus superficialis. 
Median slip of tendon of above to 

ventral skin. 
Lateral slip of same tendon. 
M. cranio-hyoideus. 
Cloaca. 
Coeliac artery. 

I M. constrictor pharyngis — first, second, 
j and third divisions. 

M. copulo-copularis. 

M. copulo-palatinus. 

Principal head 

Posterior division 

Aiiterior division 

Common portal vein. 

M. copulo-quadratus profundus. 

M. copulo-quadratus superficialis. 

Area of mucosa immediately behind the 
dental apparatus puckered by tne 
withdrawal of the latter. It flattens 
out when the teeth are protruded. 

44 



of the M. copulo-ten- 
tac u lo-coroD arius. 



342 



PROFESSOR FRANK J. COLE 



d.^esxt. 
d.t. 

d.t: 



e.l.h. 
e.l.b: 



f.r. 

g-b. 
g.i. 

h.c.g. 

h.c.t/.' 

h.c.g." 

h.c.p. 

h.g. 

h.p. 
h.p.' 

h.p." 

hy. 

hyp. 

i.b.p.' 

i.c.b. 
Lib. 
i.l.c. 

inf. 

int. 
I.l.c. 

l.lg. 
l.lg." 
l.lg." 

l.p. 
l.p.c. 

mc. 

m.e. 

mes. 

mes.' 



Ductus oesophago-cutaneus. 
Median dorsal tooth. 

Curved median bar of soft pseudo- 
cartilage, containing nodules of true 
(soft) cartilage, arising from the base 
of the median dorsal tooth, and curving 
backwards round the anterior margin 
of the palatine commissure to be 
attached to the posterior extremity of 
the subnasal bar. 

External bar of the anterior segment of 
the basal plate. 

External lateral velar bar. 

Short rod of soft cartilage connecting 
e.l.b. with the posterior extremity of 
the inferior process of the '' pterygo- 
quadrate." 

The four fenestrse of the skull. 

The so-called "fin rays" composed of 
soft cartilage. 

Gall bladder. 

Gill lamella. 

M. hyo-copulo-glossus. 

Anterior end of "tendon " of h.c.g. 

"Tether" of /(.C.J7. 

M. hyo-copulo palatinus. 

Habenular ganglion of the diencephalon. 
The shaded portion is the pineal organ. 

Hypophysial plate. 

Rod connecting h.p. with the posterior 
transverse bar of the nasal capsule. 

Hypophysial plate + the fused rod h.p.' 

" Hyoid " arch. 

Hypophysis cerebri. 

Internal bar of the anterior segment of 
the basal plate. 

Inferior chondroidal bar. 

Internal lateral velar bar. 

Inferior lateral cartilage. 

Infundibuliim cerebri. 

Intestine or mid-^ut. 

Lateral "labial" cartilage. 

M. longitudinalis linguae, 

Peripheral tendon ) j. , , 

Central tendon. ) 

Lateral plate of the nasal capsule. 

Large left posterior cardinal vein. 

Mesocoble, or mesencephalic ventricle. 

Mesencephalon. 

Median dorsal mesentery. 

Ventral fusion of rectum with body wall 
forming a rudimentary median ventral 
mesentery. 



mes. 

mt.e. 

mth. 

n.a. 

na.ph. 

nas. 

n.c. 

n.r.8 

nt. 

n.th. 

o.a. 

obi. 

oes. 

o.l. 
par. 

pc. 
p.d.p. 

p.e. 
p.e.p. 
p.e.s. 

p-g- 
p-g' 

p.h. 

p.h.d. 

p.h.v. 

pi. 

pl.c. 

pl.c! 

pi. cm. 

p.l.l. 

pn. 

pp. 

p.q. 

p.t.v.b. 

p.v. 

q.p. 

q.p.' 

red. 

r.intest.X. 

r.p.c. 

s.ao. 

s.d. 

s i.v. 

s.l.c. 

sn.b. 

sp. 

sjy.cd. 
sp.slc. 



Oblique dorso-posterior extension of 

above fusion. 
Metencephalon. 
Mouth. 

External opening of the nasal tube. 
Naso-pharyngeal or hypophysial tube. 
M. nasalis. 
Nasal capsule. 
, Eighth ring of the nasal tube (cp. Part 

HI., fig. 1). 

Chorda dorsalis. 

Nasal lube. 

Oral aperture. 

M. obliquus. 

"Oesophagus," or properly the pharyn- 
geal gut. 

Olfactory lamina. 

M. parietalis, forming the myotomes. 

" Parachordal " cartilage. 

Posterior arch of the dental plate. 

Prosencephalon. 

M. palato-ethmoidalis profundus. 

M. palato-ethmoidalis superficialis. 

Porus genitalis (abdominal pore). 

Thickened fibrous band surrounding the 
porus genitalis. 

Portal heart. 

Posterior hepatic duct. 

Posterior hepatic vein. 

"Palatine" bar. 

External head ) of the M. palato-cor- 

Internal head / onarius. 

" Palatine " commissui-e. 

Posterior lobe of the liver. 

Pronephros. 

M. perpendicularis. 

" Pterygo-quadrate." 

Posterior transverse velar bar. 

Portal or supra-intestinal vein, 

M. quadrato-palatinus. 

Tendon of g.p. 

M. rectus. 

R. intestinalis vagi. 

Smaller right posterior cardinal vein. 

Dorsal or systemic aorta. 

Segmental duct. 

Sub-intestinal vein. 

Superior lateral cartilage. 

Subnasal bar. 

Spermatic portion of the hermaphrodite 
gonad. 

Spinal cord. 

Suprapharyngeal skeleton. 



ON THE GENERAL MORPHOLOGY OF THE MYXINOID FISHES. 



343 



sp.sk: Anterior ) ^^^^necting processes of sp.sl: 
sp.sfc" Median. ) 

ss. Slime sacs — numbered from before back- 
wards. 
sy. "Sympathetic" nerve, formed by the 
fusion of the two RR. iutestinales vagi. 
thy. Thyroid vesicle. 
t.o. M. transversus oris. 
t.p. M. tentacularis posterior. 



tr. "Trabecula." 
ur.pp. Urinary papilla and apertures 
vel. Pharyngeal velum. 
v.q. Ventral division 
v.q.' Middle division 
v.q." Dorsal division 
v.qt. Ventriculus quartus 
v.s. M. velo-spinalis. 



of the M. velo-quad- 
ratus. 



Plate I. 



Fig. L Median vertical longitudinal section of a 26-|-cm. Hag. x 8^. The drawing is necessarily based 
on the study of a number of sections immediately about the median plane, and was originally sketched out 
with Edinger's projection apparatus. Care has been taken to make it as accurate as possible. Visceral 
cavities, shaded ; vascular spaces, red ; hard and soft cartilage, green ; pseudo-cartilage, obliquely striated. 
The tentacles and nasal rings are numbered from before backwards, and the isolated spaces in the anterior 
portion of the brain represent vestiges of the ventricular system. 

Fig. 2. Reconstruction from serial sections of the branchial apparatus of a 25-cm. Hag, seen from the 
ventral surface, x 9|. The sliape of the third pair of gills is somewhat exaggerated owing to the fact 
that the specimen had to be cut into three pieces to get it into the microtome, and one of the two breaks 
was through this region. Gills, afferent and efferent ducts, and the external branchial apertures displayed 
laterally, in order that all may be shown, but the position of all the structures in the longitudinal plane is, of 
course, correctly indicated. Outer and inner surfaces of gills not differentiated. Ventral or cardiac aorta, 
red. Afferent and efferent gill ducts numbered from before backwards. 



Plate II. 

Fig. 3. Reconstruction of the detached vesicules (251 in number) forming the thyroid gland of a 25-cm. 
Hag, viewed laterally from the left side. x 11. For the reason mentioned above, the length of vesicle* is 
somewhat exaggerated, which explains also the gap between sections 2560 and 2600. A few of the alveoli 
are represented displaced vertically for the sake of clearness, but only occasionally do the vesicles overlap in 
the transverse plane. Similarly the club-shaped muscle has been slightly depressed, so as not to overlap the 
anterior vesicles. Afferent gill ducts numbered from before backwards. Cardiac aorta, red ; and afferent 
branchial arteries shown cut off short. 

Fig. 4. Dissection from the ventral surface of the portal heart, gall bladder, and adjacent region of a 38-cm. 
Hag. X 3. The two lobes of the liver, with the gall bladder, have been turned to the left and pinned down. 
Hepatic ligament and cystic peritoneum (together with the proximal portion of the cystic vein) removed. 
Intestine displaced slightly to the right. Heart removed. Arteries, cross-striated. 

Fig. 5. Dissection from the right side of the cloacal region of a 37-cm. Hag. x 4. Myotomes and slime 
sacs numbered from before backwards, the former being also partly cut aM'ay. Segmental duct dissected clear 
and displaced dorsally. Sphincter cloacse, and right wall of cloaca, body cavity, and abdominal pore removed. 



Plate III. 

Fig. 6. Simplified diagrammatic horizontal section of the Myxinoid gill. x 17. The number of gill 
lamellae is greater than that shown. The upper part of the section represents the appearance of a tangential 
section, the lower part that of a central section. Afferent arteries, blue ; efferent arteries, red ; water spaces, 
shaded. 

Fig. 7. Section number 240 of the large series of a 25-cm. Hag. x 12^. This set of five sections 
is intended to illustrate the relations of the anterior region of the gut. All were drawn with Edinger's 
projection apparatus. As regards the skeleton, the sections should be compared with the charts accompanying 
Part III. of this work. Muscles indicated by circles ; vascular spaces of whatever character, red ; cartilage, 
green ; pseudo-cartilage, obliquely striated ; cavity of mouth and nasal tube, shaded. 



344 PROF. FRANK J. COLE ON THE MORPHOLOGY OF THE MYXINOID FISHES. 

Fig. 8. Section immher 390. x 12^. Colours, etc., as above. The section passes through the anterior 
process of the hypophysial plate, and through the anterior region of the anterior arch of the dental plate. 
The numbers refer to the outer row of the ventral teeth — 1, 1', 2, 3 = the anterior and posterior cusps of 
tooth 1, and teeth 2 and 3. 

Fig. 9. Section number 510. x 14. Colours, etc., as above. The section passes through the thalamus 
of the brain immediately in front of the infundibulum — -in fact, a small portion of the infundibular cavity 
appears in the section. The third, fourth, and fifth teeth of the inner row of ventral teeth, and the 
sixth, seventh, and eighth teeth of the outer row, are indicated by numbers. Cavity of naso-pharyngeal 
(liypophysial) tube, shaded. 

Plate IV. 

Fig. 10. Section number 720. x 10. Colours, etc., as above. The section passes through the pharynx 
after the latter has been joined by the naso-pharyngeal tube with its lateral diverticula and also through 
the base of the velum or pharyngeal valve which projects boldly into the pharyngeal cavity. 

Fig. 11. Section number 840. x 9^. Colours, etc., as above. The section passes through the velum 
or pharyngeal valve just anterior to the anterior transverse velar bar (cp. Part I., fig. 16, and Part III., 
fig. 2), and illustrates the relation of the valve to the cavity of the oesophagus, and also the relative 
position of the "branchial" arches. 

Fig. 12. Section number 2680. x 10. Colours, etc., as above. Eff'erent gill ducts numbered from 
before backwards. An afferent branchial artery is shown entering the third gill on the right side, 
and an efferent branchial artery leaving the fourth gill on the left side. The section passes through practically 
the centre of the gill region (cp. fig. 2), and is intended to illustrate the topographical relations of the 
thyroid vesicles. 

Fig. 13. Section number 3550. x 10. Colours, etc., as above. To illustrate the relations of the 
biliary apparatus generally. The figure is diagrammatic in one respect, since the whole of the bile duct 
does not appear in one section. Its relations are, however, correctly shown. On the left side a dorsal 
root of a spinal nerve with its ganglion are shown. On the right the section shaves the efferent duct 
of the thirty-fifth slime sac (counting from before backwards). 



■jis. Roy. 



Soc. Ediir 

Prof. F. J. Cole on the Morphology of Myxine,— Part V. Plate I. 



Vol. XLIX. 





MFarlane k Erskine. Lith. Edin- 



Icfis. Roy. 



Soc. Edin' ^<^1- XLIX. 

Prof. F. J. Cole on the Morphology of Myxine. — Part V. Plate II. 



a.p.v. 





Fig. 4. X 3 




Fig. 5. X 4 



M'Karlaae & Erjkine, Lith. EiUd. 



Is. Roy. Soc. Edin^ Vol. XLIX. 

Pkof. F. J. Cole on the Morphology or Myxtne.— Part V. Plate IIT. 




l.l.c. 



M'Farl&ne & Erskinc, Lith. Edin. 



jis. Rof. Soc. Edin' 



Vol. XLIX. 
Prof. F. J. Cole on the Morphology of Myxine.— Part V. Plate IV. 




Ml'iirlane & Eiskine. Lith. Edin 



.xSH-uc^ 




( 345 ) 



IV. — On the Skulls of Antarctic Seals: Scottish National Antarctic Expedition. 
By William S. Bruce, LL.D., Director of the Scottish Oceanographical Labora- 
tory. (With Five Plates.) 

(MS. received March 13, 1913. Read May 5, 1913. Issued separately June 27, 1913.) 

Although the osteology of Antarctic seals has been very completely discussed, 
notably by the late Dr J. E. Gkay, Sir William Turner, and Professor Robert 
Thomson, yet the literature regarding the subject is somewhat scattered. I have 
therefore considered that it might be important from the point of view, as it were, of 
an index to publish a complete series of photographs of a set of the skulls of seals 
taken by the naturalists of the Scotia during the Antarctic voyage of 1902-1904. 

The species considered are : — 

1. Leptonychotes Weddelli (Gill) : The Weddell Seal. 

2. Stenor/iynchus lejdonyx (F. Cuvier) : The Sea-leopard. 

3. Lohodon carcinoj)haga (Gray) : The Crab Eater, or White Antarctic Seal. 

4. OmmatopJwca Rossi (Gray) : The Ross Seal. 

5. Otaria jvhata (Forster) : The Patagonian Sea-lion. 

The type collection of the seals' skulls taken by the Scotia naturalists are chieHy 
housed in the Anatomical Museum of the University of Edinburgh, and in the Scottish 
Oceanographical Laboratory ; those housed in the Anatomical Museum of the Uni- 
versity of Edinburgh have been duly recorded in Sir William Turner's excellent 
descriptive catalogue entitled Marine Mammals in the Anatom,ical Museum of the 
University of Edinburgh ; while Professor Robert Thomson, of Cape Town Uni- 
versity, has contributed a paper to the Transactions of the Royal Society of Edinburgh 
entitled " Osteology of Antarctic Seals," which also appears in the Scientific Reports 
of the Voyage of S. Y. " Scotia," volume iv. 

Beyond this it is unnecessary to say more at the present time, except to refer to 
the plates published herewith. It will be observed that the skulls of the species of 
each seal are shown from every possible aspect : — 

From ( i ) The posterior aspect. 

(2) The anterior aspect. 

(3) The lateral aspect. 

(4) The inferior aspect without the lower jaw. 

(5) The inferior aspect with the lower jaw. 
(5a) The superior aspect of the lower jaw. 

(6) The superior aspect. 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 4) 45 



34(5 DR W. S. BRUCE ON THE SKULLS OP ANTARCTIC SEALS. 

The figures speak for themselves, and, as already stated, are intended to be an 
index for the use of museums and for the use of naturalists exploring in Antarctic and 
Subantarctic Regions. 

1 have considered the measurements and weights of Antarctic seals in the flesh in a 
separate paper ; the anatomy of the Weddell Seals taken by the Scotia has been fully 
considered by Professor David Hepburn and Dr Harold Axel Haig ; while Dr R. N. 
RuDMOSE Brown has dealt with " The Habits and Distribution of the Seals of the 
Weddell Sea."* 

* Vide Scie7itific Reports of the Voyage of S.Y. "Scotia," vol. iv., parts ii., iii., v., vi., ix., x., xi., xii., and xiii. 



Vans. Roy. Soc. Edin. 



Vol. XLIX. 



Bruce : " Skulls of Antarctic Seals." — Plate I. 




LEPTONYCHOTES WEDDELLT 



;rans. Roy. Soe. Edin. 



Bruce: "Skulls of Antarctic Seals." — Plate II. 



Vol. XLIX 









STENORHYNCHUS LEPTONYX. 



T 

\rmm. Roij. Sol: Edin. 



Vol. XLIX. 



Bruce: "Skulls of Antarctic Seals." — Plate III. 




LOBODON CARCINOPHAGA. 



Trans. Roy. Soc. Edin. 



VuL. XLIX. 



Bruce: "Skulls of Antarctic Seals." — Platk IV. 







OMMATOPHOCA ROSSI. 



■ans. Roy. Soc. Edin. 



Vol. XLIX. 



Bruce: "Skulls of Antarctic Seals." — Plate V. 




■t^wg^^: 



OTARIA JUBATA. 



( 347 ) 



V. — Polychseta of the Families Serpulidae and Sabellidae, collected by the Scottish 
National Antarctic Expedition. By Helen L. M, Pixell, B.Sc, F.Z.S., 
Demonstrator of Zoology and Reid Fellow, Bedford College, University 
of London. Communicated by Dr J. H. Ashworth. (With One Plate.) 

(MS. received January 22, 1913. Read February 17, 1913. Issued separately June 24, 1913.) 

For the opportunity of examining these specimens I am indebted to Dr W. S. 
Bruce, the leader of the Scottish National Antarctic Expedition, 1902-1904. I should 
like to take this opportunity of expressing my thanks to him, and also to Dr Marett 
Tims for his assistance during the preparation of the paper. 

The following is a list of the ten species, of which four are new. Eight different 
genera are represented : — 

Family Serpulidae. 

Serpula vermicularis Linnaeus, from various stations (see p. 348). 

Apomatus hroivnii n. sp.. Station 417, lat. 71° 22' S., long. 16° 34' W., 1410 fathoms. 

Salmacina dystevi (Huxley), Station 461, Gough Island, lat, 40° 20' S., long. 

9° 56' 30" W., 100 fathoms. 
Spirorhis antarcticus n. sp.. Station 325, Scotia Bay, South Orkneys, lat. 
60° 43' 42" S., long. 44° 38' 33" W., 10 fathoms. 
,, patagonicus Caullery and Mesnil, Station 478, Table Bay, Cape Town ; 

also Station 118, Port Stanley, 4 to 5 fathoms. 
,, falklandicus u. sp., Station 118, Port Stanley, Falkland Islands, 4 to 5 

fathoms. 

Family Sabellid.<e. 

1. Dusychone violacea (Schmarda), Station 478, Table Bay, S. Africa, shallow water. 

2. Eurato melanostigma (Schmarda), Station 24, Cape Verde Islands, shore. 

3. Potamilla antm'ctica Giayiev, Station 349, Tussock Island, Falkland Islands, shore. 

4. Potamis scotix n. sp.. Station 417, lat. 71° 22' S., long. 16° 34' W., 1410 
fathoms. 

Family Serpulid^. 

Genus Serpula. 
Generic characteristics : — 

1. Collar-setse of bayonet-shape, with spines at base of blade. 

2. Operculum funnel-shaped, with numerous radii ending in serrations on margin. 

3. Uncini with only a few teeth. 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 5). ' 46 



348 HELEN L. M. PIXELL ON 

Serpula vermicularis Linnseus, 1767. 

For synonyms see MoRCH (14, p. 381), Saint-Joseph (16, p. 328), Ehlers (5, p. 219) 
and Fauvel (6, p. 691). 

Specific characteristics : — 

1. Collar-setae with generally two large blunt processes at base of blade. 

2. Uncini generally have five teeth — there may be four, six, or seven. 

3. Branchiae (twenty to thirty-two pairs) long, with numerous pinnae and bare 

filamentous extremities. 

4. Serrations on operculum vary very much in number — there may be as 

many as a hundred. 

5. Maximum length recorded 50 mm., with 157 segments. 

Localities.— ^tSiiion 325, Scotia Bay, South Orkneys, lat. 60° 42' 43'' S., long. 
44° 38' 33" W.; Station 346, Burdwood Bank, lat. 54" 25' S., long. 57° 32' W., 
56 fathoms ; Station 478, Table Bay, shore; Station 461, Gough Island. Part of a 
colony, with several specimens having intertwined tubes, was taken in a trap in 
MacDougal Bay, South Orkneys. 

These tubes are cylindrical with trumpet-like ends, as described by M'Intosh 
(12, p. 516) for S. narconensis Baird. I fully agree with Ehlers (5, p. 219) that this 
species is synonymous with S. vermicularis ; the specimens show a considerable 
amount of variation, as is usual in this species, but there seem to be no points of 
difi"erence having specific value. 

There is one specimen only from Station 478, Table Bay, May 1904, with its tube 
adhering for its whole length to the shell of Mytilus. This tube is triangular in 
section, with a very distinct median ridge ending in a point overhanging the mouth. 

Three tubes from Burdwood Bank, taken in the trawl in 56 fathoms, December 1905, 
Station 346, are solitary, almost cylindrical, but enlarge very gradually towards the 
anterior end, where there is a distinctly everted peristome. The tubes are ringed with 
former peristomes at intervals, showing successive zones of growth. They have no 
doubt been attached to some substratum, for a part of their length, and are more or less 
overgrown with a bryozoon. The largest, though incomplete posteriorly, is 70 mm. 
long and 4 mm. in diameter across the peristome. One of the smaller ones contained an 
animal 25 mm. long, having an operculum with thirty-two serrations, and gills with bare 
terminal ends intermediate in length between those from Scotia Bay and Table Bay. 
Fig. 1, a and h, shows the variation amongst the collar-setae of a single fascicle : the 
former having the typical two processes ; the latter, four almost equally large ones. 

All the specimens are between 2 and 3 cm. in length, and have seventy to eighty 
segments. Their opercula have thirty-two to forty-one serrations, and their uncini 
generally six or seven teeth. The branchiae (twenty-two to twenty-eight pairs) have 
in some cases very long bare terminal ends measuring quite 1 mm., instead of the more 
usual "48 mm. according to Saint- Joseph (16, p. 330). 



POLYCH^TA OF THE FAMILIES SERPULID^ AND SABELLID^. 349 

The few specimens from Gough Island have less than thirty-two serrations to their 
opercula, but it seems that they are quite young. 

Genus Apomatus Philippi, 1 844. 

Generic characteristics : — 

1. Operculum globular, terminating a gill retaining its pinnse. 

2. Some of the thoracic setse are bladed sickles (setae of Apomatus). 

3. Terminal dorsal gland present. 

Apomatus hroivnii n. sp. 
Specific characteristics : — 

1 . Collar-setse with small fin at base of blade ; fig. 2, a. 

2. Uncini begin on the second thoracic segment, and have only seven or eight 

teeth ; fig. 2, d and e. 

3. Abdominal setse sickle-shaped ; fig. 2, c. 

Five specimens, more or less incomplete, from Station 417, lat. 71° 22' S., long. 
16° 34' W., trawled at a depth of 1410 fathoms, 18th March 1904. 

This form differs from all other previously described species of the genus in the 
characteristics 1 and 2 above, i.e. in the shape of the collar-setse and the uncini. 
There is, however, complete agreement with the two main generic characteristics (1 and 
2 above) as given by Saint- Joseph (16, p. 373). It is to be noted that this species 
comes from very deep waters. 

Tubes almost cylindrical and smooth, except for inconspicuous growth lines. The 
diameter of the oral aperture of the largest tube is 3*5 mm. ; 50 mm. from this is the 
other aperture 175 mm., but this is probably not the real end of the tube, which 
may have been attached to some substratum by a further narrower portion. Most 
of the tubes have obviously been longer, but only one, the smallest, shows a scar of 
attachment. 

The animals were preserved in their tubes. The total length of the longest is 
27 mm., and of the shortest 16 mm. ; the greatest width is 2 mm. The fifteen pairs 
of branchiae are spirally coiled. They measure from 6 to 10 mm. in length, and are thin, 
with long fine pinnse extending almost to the tops of the rachises. 

One animal has its branchial crowns projecting from the tube, and has lost its 
operculum. TVo others have lost their branchial crowns altogether. In the remaining 
two, the second dorsal gill on the left or right terminates in a transparent globular 
operculum. In one there is a secondary rudimentary operculum on the other side. 

The collar is L mm, deep, entire ventrally, but notched laterally, and very much 
folded. There are only five or six collar-setse of the characteristic form, which, with 
about half a dozen ordinary capillary setse with narrow blades, make only a small 
fascicle. The other thoracic fascicles contain numerous setse, comprising ordinary 
narrow blades and setse of the usual Apomatus type (fig. 2, h). 



350 HELEN L. M. PIXELL ON 

The abdomen has only a very short asetigerous region, followed by 80 to 110 narrow 
segments with sickle-shaped setfe (fig. 2, c). Many of these sickles have, however, 
become much straightened, owing, no doubt, to their prolonged immersion in alcohol. 
The posterior setae are very elongated. The dorsal gland, generally present in the 
young, is very little developed, and in two specimens appears to be wanting. There 
is a deep ventral fsecal groove which turns to the right on reaching the thorax. 

The uncini have only seven or eight sharp teeth (fig. 2, d and e) ; consequently, if 
Saint- Joseph's classification (16, p. 263) according to the characteristics of the uncini 
were rigorously adhered to, this species would have to be placed in a new genus in 
Group I., with Serpula, etc., instead of in the genus Apomatus, which Saint- Joseph 
includes in his Group V., which is made up of forms whose uncini have very numerous 
and fine teeth. 

Genus Filograna Oken, 1815. 
Generic characteristics : — 

1. Tube very slender, filiform, colonial. 

2. Branchiae eight. 

3. Thorax with seven to nine segments. 

4. Collar-setae with a large fin-like expansion at base of blade. 

5. Other thoracic setae sickle- shaped (setae of Salmacina), or ordinary bladed 

ones. 

6. Abdominal setae more or less geniculate and serrated. 

7. Hermaphrodite. 

Sub-genus Salmacina Claparede, 1870 (2, p. 176). 

1. No operculum. 

2. The ends of the branchiae may or may not have spathulate enlargements due 

to the presence of large granular cells. 

Salmacina dysteri (Huxley), 1855. 

Protula dysteri Huxley, 1855 (10, p. 113). 
FiloporafiJograna Dalyell, 1853 (3, p. 250). 

Specific characteristics : — 

1. Branchiae with spathulate enlargements containing granular masses at their 

ends. Similar granules occur at the ends of the pinnae, and just in front 
of their bases along the gill rachises. 

2. Spermatozoa developed in segments anterior to those producing ova. 
Zocoi%.— Station 461, Gough Island, lat. 40° 20' S., long. 9° 56' 30" W., 

100 fathoms. Several fairly large masses brought up by the trawl with Serpula 
vermicularis, coral, etc. No specimens in process of budding were observed. 



POLYCH^TA OF THE FAMILIES SERPULID^ AND SABELLID^. 351 



Genus S'pirorhis Lamarck, 1801. 

Characteristics of genus and sub-genera have been given in full (15, pp. 

792-799). 

Sub-genus Paralxospira Caullery and Mesnil (1, p. 202), emend. Pixell (15, pp. 795 

and 799). 

Spirorhis antarcticus n. sp. 

Specific characteristics : — 

1. Collar-setae — a few small finely striated simple blades, fig, "6, c. 

2. Third thoracic fascicle contains some bladed sickle-shaped setae. 

3. Abdominal setae, flattened trumpets with one side produced. 

4. Tube triangular in section and geneially very regularly coiled (fig. 3, a). 
Locality. — Several specimens growing on fucus, dredged at Station 325, April 1903, 

Scotia Bay, South Orkneys; lat. 60° 43' 42" S., long. 44° 38' 33'' W., in 9 to 10 
fathoms. 

The large thick tubes are almost perfectly triangular in section and very regularly 
coiled to form a disc, with concave upper surface, and measuring up to 4 mm. across 
(fig. 3, a). There are nearly always young tubes attached to them, one specimen bearing 
as many as twelve of different sizes. In this they resemble Caullery and Mesnil's 
(l, p. 203) Spirorbis aggregatvs. 

The opercular plate may be flat, convex, or conical, but has in every case a thin 
flattened talon. Small circular perforations occur in the calcareous plate, and through 
these project small membranous projections, which are generally thorn-shaped. Though 
the opercula vary so much, there is no doubt that all the variations should be included in 
the same species, for there are intermediate forms between the thin, flattened plate, 
which seems to be characteristic of young specimens, and the conical form. It is also 
very questionable whether details as to the shape of the operculum are ever suflficiently 
constant to be of much value in specific determination. 

Many of the specimens are of very large size, some measuring over a centimetre in 
length. There are nine long branchiae : the second on the left is transformed into the 
pedicle of the operculum, and is slightly larger than the others. 

This species could only be included in the sub-genus Paralwospira, provided that 
the modified characteristics as suggested in a previous paper (15) be adopted for it, 
owing to the fact that it differs from all previously described Paralseospira in 
having simple blades for the collar-setae (fig. 3, c). These are remarkably small, and 
there are only seven on each side, so that the whole fascicle is very insignificant. 
The other thoracic fascicles are much larger, and the uncini are about 40 mm. 
long. There are about twenty abdominal segments following a rather long aseti- 
gerous region. 



352 HELEN L. M. PIXELL ON 

Spirorhis patagonicns (Caullery and Mesnil), 1897 (I, p. 205); Ehlers, 1901 

(5, p. 224). 
Specific characteristics : — 

1. Talon of operculum, a prolongation of the terminal shallow funnel. 

2. Collar-se-tse small, with a fin-like expansion at the base of finely serrated 

blade (fig. 4, a). 

3. Numerous sickle-shaped setse to 3rd thoracic segment. 

4. Abdominal setae large, flattened, and trumpet-shaped. 

5. Large somewhat irregularly coiled tubes. 

Localities. — Several specimens on the shell of Mytilus, from Station 478, Table Bay, 
Cape Town Docks, May 1904. Other specimens, with Spirorhis falldandicus, on a 
stone from the shore at Station 118, Port Stanley, Falkland Islands, January 1903. 

Most of the tubes are rather smaller than those described by Caullery and Mesnil. 
They are generally loosely and irregularly coiled. All ages seem to be represented, and 
both the young with three thoracic segments only and the adults agree with the 
descriptions given by these authors. 

There are apparently ten branchige, including the opercular pedicle, and about 
twenty abdominal segments. The largest specimen obtained whole from its tube 
measured 6 mm. 

Paradexiospira Caullery and Mesnil (1, p. 195), emend. Pixell (15, p. 793). 

Spirorhis falklandicus, n. sp. 

1. Last thoracic segment without dorsal setae. 

2. Collar-setae small, very finely serrated blades with large fin at base. 

3. Operculum very similar to that of ^S". vitreus. 

4. The 3rd thoracic fascicle contains bladed sickles (setae of Apomatus) (fig. 5, c). 
Several specimens, with S. patagonicus, on a stone from the shore at Station 118, 

Port Stanley, Falkland Islands, January 1903. 

This is the first time that a species of the sub-genus Paradexiospira has been 
recorded from the Southern Hemisphere. Caullery and Mesnil (1, p. 195) state that 
the three species known to them, namely S. cancellatus Fabr., S. vitreus Fabr., and 
S. violaceus Lev., inhabit Greenland and Iceland. S. vitreus has since been found in 
the North Pacific (15, p. 793). 

The tubes of Spirorhis falklandicus generally have three longitudinal ridges, the 
median one sometimes ending in a projection overhanging the mouth of the tube. The 
terminal part may be more or less ascending, and is rather more than 1 mm. in dia- 
meter. Only about one and a half turns of the spiral can be seen from the top, and the 
tube measures 3 mm. across. 

Branchiae ten, the second on the right bearing the operculum, as usual. In young 
specimens the operculum is almost flattened on the top, and has a massive, sometimes 



POLYCH^TA OF THE FAMILIES SERPULID^ AND SABELLID^. 353 

bifid, talon (fig. 5, a). In older specimens this is replaced, as in S. vitreus, by a thin 
funnel-shaped one without a talon ; in one animal examined, both were present, one 
inside the other, the thin funnel-shaped one having been produced before the loss of 
the smaller massive one. 

There is a deep collar, and the species differs from S. vitreus and all other previously 
described Paradexiospira in having collar-setae with such fine serrations that it is 
almost impossible to recognise any at all. I have already suggested that the character- 
isation of Caullery and Mesnil's sub-genera should be made more general (15, 
p. 793), and this is another instance of the necessity of such a step, for S. falk- 
landicus is undoubtedly closely related to the three forms known to Caullery and 
Mesnil and placed by them in the sub-genus Paradexiosfira. Of these collar-setae 
there are about nine, making with a few of the usual fine capillary ones a bundle con- 
siderably smaller than the other thoracic fascicles. 

The uncini of the thorax are 40 to 50 mm. long, and contain about seventeen teeth ; 
those of the abdomen are much smaller, being only 20 mm. long. 

The abdominal setae are flattened and trumpet-shaped, with one side produced. 

There seem to be twenty or more abdominal segments, but I did not succeed in 
obtaining a complete specimen from a tube, the state of preservation not being good. 

Family Sabellid^. 

Genus Dasychone Sars, 1861. 

1. Thoracic and abdominal tori with single rows of avicular crotchets. 

2. Thoracic setae of one kind only. 

3. Branchiae with dorsal appendices. 

Dasychone violacea {^cXwaax^Q.), 1861. 

Sabella violacea Schmarda (17, p. 34). 
Da&ychone violacea M'Intosh (12, p. 504). 

1. Dorsal appendices large and somewhat spathulate, covering an ocellus on 

either side of the rachis. 

2. A dark pigment spot on each side of every segment, between the fascicles of 

setae and the tori. 

3. Thorax of eight setigerous segments. 

4. Crotchets with very fine serrations above the large fang. 

Locality. — One specimen from Station 478, Table Bay, May 1904. Total length 
26 mm., of which the branchiae form 8 mm. ; the width is about 4 mm., except close to 
the posterior end, where it decreases rapidly. There are seventy abdominal segments. 

The setae and crotchets agree entirely with those so clearly described and figured 
by M'Intosh. 



354 HELEN L. M. PIXELL ON 

It will be noticed that this specimen is smaller than those described before : it 
agrees more nearly with Schmarda's specimens in having only twenty branchiae. Each 
of these has seven pairs of well-formed dorsal appendices, with rudiments of others towards 
the tips. The specimen was by no means fully developed, nor was it sexually mature. 

No trace of colour remains, except on the gills and the pigment spots already alluded 
to. The tube is also missing. 

Genus Eurato Saint- Joseph, 1894 (16, p. 249). 

Generic characteristics :— - 

1. Thoracic and abdominal tori with a single row of avicular crotchets. 

2. Branchiae without dorsal appendices and arranged in a single series. 

3. Thoracic setae all simple blades. 

Eurato melanosfigma (Schmarda). 
Sahella melanostigma Schmarda, 1861 (17, p. 36), from Jamaica. 

1. Branchiae about half the length of the body, few in number, with short pinnae 

and an interbrancliial membrane. 

2. A pigment spot on each side of every segment, between the fascicles of seta3 

and the tori. 

3. Thorax of five setigerous segments (? always). 

Several specimens from Station 24, Porte Grande, St Vincent, Cape Verde Islands. 

These specimens seem to be immature, and are much smaller than Schmarda's ; but 
from the latter's brief description they would seem to be the same species. 

They are 5 mm. or less in length, the branchiae when intact making up nearly 2 mm. 
of this ; the greatest width is about 1 mm., and the largest number of abdominal seg- 
ments counted was thirty-five. 

The collar has two pointed ventral lobes, but is otherwise not very high, and stops 
short some distance from the mid-dorsal line. 

The collar-setae have long narrow blades (fig. 6, a) ; there are only four other setiger- 
ous thoracic segments, and these have slightly wider blades in addition to the long 
narrow ones (fig. 6, h). The abdominal segments have the wider blades only. The 
thoracic tori have about twelve avicular crotchets (fig. 6, c) ; the abdominal only seven. 

The anal segment is asetigerous and very distinctly bi-lobed. 

There are only seven or eight pairs of gills. 

In no specimen were gonads apparent ; this immature state may account for the 
small number of branchiae and thoracic segments. 

One very young specimen was 2 mm. long and 1 mm, wide, only very slightly 
tapered at each end, and containing eighteen setigerous segments ; all of these seemed 
to have the setae ventral to the tori, i.e. to be abdominal segments, except one or two 



POLYCH^TA OF THE FAMILIES SERPULID^ AND SABELLID^. 355 

very small crowded segments anteriorly, in which no tori could be distinguished. The 
branchial crown was represented by two small buds with four finger-like projections. 

No tube was present, but to one specimen was adhering a minute portion of a 
rather thin membrane, to which very small sand-grains were fixed, and this may have 
been a portion of a tube. 

With these also was a larger specimen (8 mm. total length), in which no pigment 
spots were present, but which otherwise seemed identical with this species. 

The pigment spots in Eurato (Sahella) manicata (9, p. 255) are differently 
arranged. 

This is apparently the first time that this species has been recorded since Schmarda 
described it from Jamaica in 1861. 

Genus Potamilla Malmgren, 1865 (13, pp. 401 and 402). 

Generic characteristics : — 

1. Thoracic tori, double rows, consisting of avicular crotchets and pennoned 

setse respectively. 

2. Abdominal tori, single rows of avicular crotchets. 

3. Collar present. 

4. Thoracic setse of two forms—dorsally a few with simple blades, ventrally 

spathulo-mucronate. 

5. Abdominal setse simple blades. 

6. Branchial crown not markedly spiral, and without sub-terminal eyes on 

rachises. 

Potamilla antarctica Gravier, 1906 (7, p. 59). 

1. Branchiae seventeen pairs, with no trace of eyes, about one-fifth the length 

of the body. 

2. Avicular crotchets of thorax and abdomen of characteristic shape (see Gravier 

(8) and (7) for figures). 

Localities. — One specimen from shore pool at Station 349, Tussock Island, Falkland 
Islands ; seven incomplete specimens, all of which had lost their branchial crowns, from 
the shore at Station 118, Cape Pembroke, Falkland Islands. 

The tubes of the latter specimens are much overgrown by a colonial tunicate. 

The anterior part of the tube of the Tussock Island specimen contained numerous 
larvae, 5 mm. long, and having five or six segments. The specimen in the tube was 
well preserved, and was 31 mm. in length, of which the branchiae made up 6 mm. 

The other specimens are, without their branchial crowns, 27 mm. long and 3 mm. 
wide, and have nearly parallel sides — only a few of the terminal segments being 
narrower. The thorax consists of eight setigerous segments, and the abdomen of 
about fifty. The setse were in nearly every segment broken off below the blade, and 

TRANS. ROY. SOC. EDIN., VOL. XLIX. PART II. (NO. 5.) 47 



356 HELEN L. M. PIXELL ON 

it was only with considerable difficulty that any perfect ones were obtained. The 
collar has small dorsal lobes, and becomes gradually higher towards the mid-ventral 
line, where there is a deep fissure. 

The ventral gland-shields are very distinct, separated by deep segmental fissures, 
and divided in the abdominal region by a deep and narrow faecal groove, which crosses 
obliquely to the right on reaching the thorax, and is continued as a very shallow groove 
alons the mid-dorsal surface. 

No pigment spots could be made out at the posterior end, and there is practically 
no colour in the preserved animals ; only faint transverse markings could be made 
out on the branchiae. In all other points there seems to be exact similarity with 
Gravier's widely distributed sjDecies. At the same time, I cannot help thinking that 
the variation, which he points out (8, p. 145) as existing in two of the characteristics 
on which he has based his species in differentiating it from P. neglecta Malmgren and 
P. incerta Lang, tends to do away with its distinctness. 

Genus Potamis Ehlers, 1887(4, p. 278). 

Generic characteristics : — 

1. Thoracic tori double rows : avicular crotchets with long bases and pennoned 

setse respectively. 

2. Abdominal tori single rows of avicular crotchets. 

3. Collar present. 

4. Thoracic setae of two forms — superior ones with rather wide blades, inferior 

sub-spathulate. 

5. Abdominal setae of two kinds — long narrow-bladed ones and short forms 

with wider blades (lanceolate) (fig. 7,f). 

6. Gill bases sub-involute. Branchiae without ocelli. 

Potamis scotise n. sp. 

1. Collar with high ventro- lateral lobes and deeply divided in the middle line. 

2. Setae and crotchets of characteristic shapes (fig. 7, a-h), diff'ering in many 

ways from those of P. spathiferiis Ehlers. 

3. Ventral gland-shields quite distinct, divided in the abdominal region by a 

wide and shallow faecal groove. 

Locality. — One specimen, without its tube, taken in the trawl at Station 417, 
lat. 71° 22' S., long. 16° 34' W., from a depth of 1410 fathoms. 

Total length 133 mm., of which 63 mm. is made up by the branchial crown, 10 mm. 
by the thorax, and 60 mm. by the abdomen. Greatest width 5 mm. 

After being preserved in dilute formalin the general colour of the specimen (is 
greenish-brown. The segments are clearly marked off from one another by inter- 
segmental constrictions, and the setae protrude to a considerable distance. 



POLYCH^TA OF THE FAMILIES SERPULID^ AND SABELLID^. 357 

The thorax has eight setigerous segments, the abdomen 52 ; the posterior abdominal 
seo-ments are much shorter antero-posteriorly, and rather wider than the front ones. 
The gill rachises are nearly 1 mm. in diameter and very nearly cylindrical ; in some 
cases they end in a small finger-like process, but the thin straggling pinnae extend at 
irregular intervals nearly up to this end. There are sixteen on the left and fifteen 
on the right, the four ventral ones on each side being much smaller. To the most 
ventral one on each side is attached the membrane which, just below, enlarges greatly 
and forms a kind of pocket on each side ; the membranes then approach one another 
and join together at the posterior end of the fissure between the ventro-lateral lobes 
of the collar. 

From the dorsal aspect there is great resemblance to Ehlers' P. spathiferus (4, 
p. 278), the collars of both species being similar. 

The first segment has no torus, and its setae are of one kind only, similar to the 
superior ones of the other segments (fig. 7, g). 

December 23, 1912. 



REFERENCES. 

(1) Caullbry, M., et Mesnil, F., "Etudes sur la morphologie comparee et la phylogenie des especes cliez 
les Spirorbis," Bull. Set. France et Belgique, xxx. pp. 185-233, Paris, 1897. 
• (2) Claparedb, E., "Annelides clietopodes du Golfe de Naples (Supplement)," ilfem. Soc. Phys. Geneve, 
XX., 1870. 

(3) Dalyell, J. G., The Power of the Creator displayed in the Creation, etc., vol. ii. p. 250, 1853. 

(4) Ehlers, E., "Florida Annelida," Mem. Mus. Comp. Zool. Harvard, vol. xv., 1887. 

(5) "Die Polychaeten des magellanischen und chilenischen Strandes," Festschr. K. Ges. Wiss. 

Gottingen, 1901. 

(6) Fauvel, p., " Sur quelques Serpuliens de la Manche et de la Mediterranee (Serpula vermicularis et 

Protula tuhularia Mont.)," Compt. Rend. Ass. Fr. Avanc. Set., Paris, 1909. 

(7) Gravier, C, Expedition Antarctique Fran^aise, 1903-05, Annelides Polych., p. 59, Paris, 1906. 

(8) Deuxieme Expedition Antarctique Franr^aise (1908-10) : Annelides Polychetes, Paris, 1911. 

(9) Grube, A. E., " Annulata Semperiana," Mem. Acad. Imper. Sci. St Petersh., 7** serie, xxv., 1878. 

(10) Huxley, T. H., "On a Hermaphrodite and Fissiparous Species of Tubicolar Annelid," Ed in. Neio Phil. 

Journ., vol. i.. No. 1, 1855. 

(11) Langerhans, P., "Die Wurmfauna von Madeira," Zeits. loiss. Zool., xxxii.-xxxiv., Leipzig, 1880. 

(12) M'Intosh, W. C, " Polychaeta," ''Challenger" Reports, Zool., xii., London, 1885. 

(13) Malmgren, a. J., "Nordiska Hafs-Annulater," Ofvers af K. Sv. Vet.-Akad. Fork., pp. 401, 402, 

Stockholm, 1865. 

(14) Morch, Otto, A. L., "Revisio critica Serp., etc.," Naturhist. Tidsskrift, i. pp. 347-470, Copenhagen, 

1863. 

(15) PiXBLL, H. L. M., "Polychaeta from, the Pacific Coast of North America : L, Serpulidaj, with a Revised 

Classification of Genus Spirorbis," Proc. Zool. Soc. Land., pp. 784-805, 1912. 

(16) Saint-Joseph, Baron de, "Les Annelides Polychetes des Cotes de Dinard," Ami. Sci. Nat. Zool. (7), 

xvii., Paris, 1894. 

(17) Schmaeda, Ludwig K., Neue tvirbellose Thiere, I. ii., Leipzig, 1861. 



358 POLYCH^TA OF THE FAMILIES SERPULID^ AND SABELLID^. 



EXPLANATION OF THE PLATE. 

Fig. 1. Serpula vermicularis L. Collar-set?e from a single specimen to show variation: la, with the 
normal two processes at base of blade ; Ih, with four nearly equal processes. x 300. 

Fig. 2. Apomatus brownii n. sp. 2a, collar-seta ; 26, bladed sickle from thorax ; '2c, sickle-shaped seta 
from abdomen ; '2d, thoracic uncinus ; 2e, abdominal uncinus. x 375. 

Fig. 3. Spirorbis antarcticus a. sp. 3a, tube with three small ones towards its centre, xl2; 3b, 
operculum, x 48 ; 3c, collar-seta, x 375. 

Fig. 4. Spirurbis patagonicus CauUery and Mesnil. Collar-seta, x 375. 

Fig. 5. Spirorbis falMandicus n. sp. 5a, operculum of young specimen, x 36 ; 56, collar-seta, x 375 ; 
be, seta of bladed sickle type, from third thoracic fascicle, x 375. 

Fig. 6. Eurato melanostigma (Schmarda). 6a and 66, thoracic setae, x 220 ; 6f, thoracic crotchet, 
x220. 

Fig. 7. Potamis scotix n. sp. 7a, thoracic avicular crotchet, x 48 ; 76, head of same, x220; 7c, thoracic 
pennoned seta, x 48 ; 7d, head of same, x220; 7e, abdominal avicular crotchet, x220; If, abdominal 
lanceolate seta, x 150; 7g, superior thoracic seta, x 150; 7/i, inferior thoracic seta, x 150. 



Trans. Roy. Soc. Edin' 



Vol. XLIX. 



PixELL : "Scotia" PoLYCHAETA. 




MTarlane & Erskine. Lith.Edin 



^sH (V'USe^- 




( 359 ) 



Vr.— Oaradocian Oystidea from Girvan. By F. A. Bather, M.A., D.Sc, F.R.S. 
Communicated hy Dr Horne, F.R.S. (With Six Plates.) [Published by per- 
mission of the Trustees of the British Museum.] 

(Read May 13, 1912. MS. received March 8, 1913. Issued separately July 24, 1913.) 



A. Introduction 

Geological Distribution in Girvan 
The Starfish Bed 

B. Systematic Description — 

Order Ami)horidea 

(Discussion of the " Carpoidea") 
Suborder Heterostelea 

(Taxonomy of the Suborder) 
Fam. Dendrocystidae . 
(Rhipidocystidae ?) 
Dendrocystis 

D. Barrandei n. sp. 
D. Sedgwicki 
D. scotica n. sp. 
D. rossica 
D. paradoxica 
Fam. Cothurnocystidae 
Cothurnocystis . 
G. Elizae n. sp. 
G. curvata n. sp. . 
Organisation and Mode of Life 
Affinities of the Genus . 
Order Rhombifera 

(Taxonomy of the Order) . 
Superfam. Glyptocystidea 
Fam. Cheirocrinidae 
Cheirocrinus 

C constrictus n. sp. 
G. interruptus 

(The names of 



CONTENTS. 

SECTION 
1 

4 



13 

16 

21 

26 

33 

34 

40 

89 

108 

127 

150 

155 

160 

161 

170 

208 

223 

245 

261 

262 

282 

283 

286 

310 

336 



SECTION 

Pleurocystis 348 

The Species and their Distribution 390 

Diagnostic Characters of the Species 394 

P. squamosa 403 

P. squamosa var. robusta . . 408 

P. filitexta 414 

P. elegans 420 

P. exornata 426 

P. anticostensis .... 432 

P. mercerensis .... 438 

P. Rugeri 444 

P. anglica 464 

P. procera n. sp 486 

P. quadrata n. sp. ... 501 

P. gibba n. sp 520 

P. foriolus n, sp 535 

Key to Species .... 551 
Relations of British to American 

Species 553 

C. General Conclusions — 

I. Distribution 559 

II. Anatomy 569 

III. Taxonomy 576 

IV. Morphology and Bionomics . . . 578 

D. List op Papers and Works referred to . 615 



E. Explanation of Plates I.-VI. 

F. Index. 

species found in Girvan are in italics.) 



A. INTRODUCTION. 

§ 1. The Scottish specimens described in this memoir are all the property of 
Mrs Robert Gray, and I am exceedingly indebted to her, not only for entrusting them 
to me for description, but for allowing them to remain in my hands for a lengthy 
period. None of the collections of the Geological Survey, nor those of Glasgow 
University, contain any cystids from Girvan. Besides her Cystidea, Mrs Gray has 
submitted to me her Crinoidea and several interesting specimens of Edrioasteroidea. 
For the present we are concerned solely with representatives of the Class Cystidea as 

TRANS. ROY. SOG. EDIK, VOL. XLIX. PART II. (NO. 6). 48 



360 r>R F. A. BATHER. 

defined by me in Lankester's " Treatise on Zoology," 1 900.* This, while excluding the 
Edrioasteroidea, includes forms that have been removed from the Cystidea, by Haeckel 
as Amphoridea, and by Jaekel as Carpoidea. 

§ 2. Though written in order to describe Girvan material, the memoir has grown to 
little less than a mon